index
int64
0
0
repo_id
stringlengths
9
205
file_path
stringlengths
31
246
content
stringlengths
1
12.2M
__index_level_0__
int64
0
10k
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/orchestration/DagManager.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.orchestration; import com.codahale.metrics.Meter; import com.codahale.metrics.Timer; import com.google.common.base.Joiner; import com.google.common.base.Optional; import com.google.common.base.Preconditions; import com.google.common.collect.Lists; import com.google.common.collect.Maps; import com.google.common.eventbus.Subscribe; import com.google.common.util.concurrent.AbstractIdleService; import com.google.inject.Inject; import com.google.inject.Singleton; import com.typesafe.config.Config; import com.typesafe.config.ConfigException; import com.typesafe.config.ConfigFactory; import java.io.IOException; import java.net.URI; import java.net.URISyntaxException; import java.util.ArrayList; import java.util.Collection; import java.util.Collections; import java.util.HashMap; import java.util.HashSet; import java.util.Iterator; import java.util.LinkedList; import java.util.List; import java.util.Map; import java.util.Properties; import java.util.Set; import java.util.concurrent.BlockingQueue; import java.util.concurrent.ExecutionException; import java.util.concurrent.Executors; import java.util.concurrent.Future; import java.util.concurrent.LinkedBlockingDeque; import java.util.concurrent.ScheduledExecutorService; import java.util.concurrent.TimeUnit; import java.util.concurrent.atomic.AtomicLong; import lombok.EqualsAndHashCode; import lombok.Getter; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.metrics.ContextAwareGauge; import org.apache.gobblin.metrics.MetricContext; import org.apache.gobblin.metrics.ServiceMetricNames; import org.apache.gobblin.metrics.event.EventSubmitter; import org.apache.gobblin.metrics.event.TimingEvent; import org.apache.gobblin.runtime.api.DagActionStore; import org.apache.gobblin.runtime.api.FlowSpec; import org.apache.gobblin.runtime.api.JobSpec; import org.apache.gobblin.runtime.api.Spec; import org.apache.gobblin.runtime.api.SpecNotFoundException; import org.apache.gobblin.runtime.api.SpecProducer; import org.apache.gobblin.runtime.api.TopologySpec; import org.apache.gobblin.runtime.spec_catalog.FlowCatalog; import org.apache.gobblin.service.ExecutionStatus; import org.apache.gobblin.service.FlowId; import org.apache.gobblin.service.ServiceConfigKeys; import org.apache.gobblin.service.modules.flow.SpecCompiler; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.flowgraph.Dag.DagNode; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.service.modules.utils.FlowCompilationValidationHelper; import org.apache.gobblin.service.modules.utils.SharedFlowMetricsSingleton; import org.apache.gobblin.service.monitoring.FlowStatusGenerator; import org.apache.gobblin.service.monitoring.JobStatus; import org.apache.gobblin.service.monitoring.JobStatusRetriever; import org.apache.gobblin.service.monitoring.KillFlowEvent; import org.apache.gobblin.service.monitoring.ResumeFlowEvent; import org.apache.gobblin.util.ConfigUtils; import org.apache.gobblin.util.reflection.GobblinConstructorUtils; import static org.apache.gobblin.service.ExecutionStatus.*; /** * This class implements a manager to manage the life cycle of a {@link Dag}. A {@link Dag} is submitted to the * {@link DagManager} by the {@link Orchestrator#orchestrate(Spec)} method. On receiving a {@link Dag}, the * {@link DagManager} first persists the {@link Dag} to the {@link DagStateStore}, and then submits it to the specific * {@link DagManagerThread}'s {@link BlockingQueue} based on the flowExecutionId of the Flow. * This guarantees that each {@link Dag} received by the {@link DagManager} can be recovered in case of a leadership * change or service restart. * * The implementation of the {@link DagManager} is multi-threaded. Each {@link DagManagerThread} polls the * {@link BlockingQueue} for new Dag submissions at fixed intervals. It deques any newly submitted Dags and coordinates * the execution of individual jobs in the Dag. The coordination logic involves polling the {@link JobStatus}es of running * jobs. Upon completion of a job, it will either schedule the next job in the Dag (on SUCCESS) or mark the Dag as failed * (on FAILURE). Upon completion of a Dag execution, it will perform the required clean up actions. * * For deleteSpec/cancellation requests for a flow URI, {@link DagManager} finds out the flowExecutionId using * {@link JobStatusRetriever}, and forwards the request to the {@link DagManagerThread} which handled the addSpec request * for this flow. We need separate {@link BlockingQueue}s for each {@link DagManagerThread} because * cancellation needs the information which is stored only in the same {@link DagManagerThread}. * * The {@link DagManager} is active only in the leader mode. To ensure, each {@link Dag} managed by a {@link DagManager} is * checkpointed to a persistent location. On start up or leadership change, * the {@link DagManager} loads all the checkpointed {@link Dag}s and adds them to the {@link BlockingQueue}. */ @Alpha @Slf4j @Singleton public class DagManager extends AbstractIdleService { public static final String DEFAULT_FLOW_FAILURE_OPTION = FailureOption.FINISH_ALL_POSSIBLE.name(); public static final String DAG_MANAGER_PREFIX = "gobblin.service.dagManager."; private static final Integer DEFAULT_JOB_STATUS_POLLING_INTERVAL = 10; public static final Integer DEFAULT_NUM_THREADS = 3; private static final Integer TERMINATION_TIMEOUT = 30; public static final String NUM_THREADS_KEY = DAG_MANAGER_PREFIX + "numThreads"; public static final String JOB_STATUS_POLLING_INTERVAL_KEY = DAG_MANAGER_PREFIX + "pollingInterval"; private static final String DAG_STATESTORE_CLASS_KEY = DAG_MANAGER_PREFIX + "dagStateStoreClass"; private static final String FAILED_DAG_STATESTORE_PREFIX = "failedDagStateStore"; private static final String FAILED_DAG_RETENTION_TIME_UNIT = FAILED_DAG_STATESTORE_PREFIX + ".retention.timeUnit"; private static final String DEFAULT_FAILED_DAG_RETENTION_TIME_UNIT = "DAYS"; private static final String FAILED_DAG_RETENTION_TIME = FAILED_DAG_STATESTORE_PREFIX + ".retention.time"; private static final long DEFAULT_FAILED_DAG_RETENTION_TIME = 7L; // Re-emit the final flow status if not detected within 5 minutes private static final long DAG_FLOW_STATUS_TOLERANCE_TIME_MILLIS = TimeUnit.MINUTES.toMillis(5); public static final String FAILED_DAG_POLLING_INTERVAL = FAILED_DAG_STATESTORE_PREFIX + ".retention.pollingIntervalMinutes"; public static final Integer DEFAULT_FAILED_DAG_POLLING_INTERVAL = 60; public static final String DAG_MANAGER_HEARTBEAT = ServiceMetricNames.GOBBLIN_SERVICE_PREFIX_WITH_DELIMITER + "dagManager.heartbeat-%s"; // Default job start SLA time if configured, measured in minutes. Default is 10 minutes private static final String JOB_START_SLA_TIME = DAG_MANAGER_PREFIX + ConfigurationKeys.GOBBLIN_JOB_START_SLA_TIME; private static final String JOB_START_SLA_UNITS = DAG_MANAGER_PREFIX + ConfigurationKeys.GOBBLIN_JOB_START_SLA_TIME_UNIT; private static final int MAX_HOUSEKEEPING_THREAD_DELAY = 180; private static final int INITIAL_HOUSEKEEPING_THREAD_DELAY = 2; /** * Action to be performed on a {@link Dag}, in case of a job failure. Currently, we allow 2 modes: * <ul> * <li> FINISH_RUNNING, which allows currently running jobs to finish.</li> * <li> FINISH_ALL_POSSIBLE, which allows every possible job in the Dag to finish, as long as all the dependencies * of the job are successful.</li> * </ul> */ public enum FailureOption { FINISH_RUNNING("FINISH_RUNNING"), CANCEL("CANCEL"), FINISH_ALL_POSSIBLE("FINISH_ALL_POSSIBLE"); private final String failureOption; FailureOption(final String failureOption) { this.failureOption = failureOption; } @Override public String toString() { return this.failureOption; } } @Getter @EqualsAndHashCode public static class DagId { String flowGroup; String flowName; String flowExecutionId; public DagId(String flowGroup, String flowName, String flowExecutionId) { this.flowGroup = flowGroup; this.flowName = flowName; this.flowExecutionId = flowExecutionId; } @Override public String toString() { return Joiner.on("_").join(flowGroup, flowName, flowExecutionId); } } private final BlockingQueue<Dag<JobExecutionPlan>>[] runQueue; private final BlockingQueue<DagId>[] cancelQueue; private final BlockingQueue<DagId>[] resumeQueue; DagManagerThread[] dagManagerThreads; private final ScheduledExecutorService scheduledExecutorPool; private final boolean instrumentationEnabled; private DagStateStore dagStateStore; private Map<URI, TopologySpec> topologySpecMap; private int houseKeepingThreadInitialDelay = INITIAL_HOUSEKEEPING_THREAD_DELAY; @Getter private ScheduledExecutorService houseKeepingThreadPool; @Getter private final Integer numThreads; private final Integer pollingInterval; private final Integer retentionPollingInterval; protected final Long defaultJobStartSlaTimeMillis; @Getter private final JobStatusRetriever jobStatusRetriever; private final FlowStatusGenerator flowStatusGenerator; private final UserQuotaManager quotaManager; private final SpecCompiler specCompiler; private final boolean isFlowConcurrencyEnabled; private final FlowCatalog flowCatalog; private final FlowCompilationValidationHelper flowCompilationValidationHelper; private final Config config; private final Optional<EventSubmitter> eventSubmitter; private final long failedDagRetentionTime; private final DagManagerMetrics dagManagerMetrics; @Inject(optional=true) protected Optional<DagActionStore> dagActionStore; private volatile boolean isActive = false; public DagManager(Config config, JobStatusRetriever jobStatusRetriever, SharedFlowMetricsSingleton sharedFlowMetricsSingleton, FlowStatusGenerator flowStatusGenerator, FlowCatalog flowCatalog, boolean instrumentationEnabled) { this.config = config; this.numThreads = ConfigUtils.getInt(config, NUM_THREADS_KEY, DEFAULT_NUM_THREADS); this.runQueue = (BlockingQueue<Dag<JobExecutionPlan>>[]) initializeDagQueue(this.numThreads); this.cancelQueue = (BlockingQueue<DagId>[]) initializeDagQueue(this.numThreads); this.resumeQueue = (BlockingQueue<DagId>[]) initializeDagQueue(this.numThreads); this.scheduledExecutorPool = Executors.newScheduledThreadPool(numThreads); this.pollingInterval = ConfigUtils.getInt(config, JOB_STATUS_POLLING_INTERVAL_KEY, DEFAULT_JOB_STATUS_POLLING_INTERVAL); this.retentionPollingInterval = ConfigUtils.getInt(config, FAILED_DAG_POLLING_INTERVAL, DEFAULT_FAILED_DAG_POLLING_INTERVAL); this.instrumentationEnabled = instrumentationEnabled; MetricContext metricContext = null; if (instrumentationEnabled) { metricContext = Instrumented.getMetricContext(ConfigUtils.configToState(ConfigFactory.empty()), getClass()); this.eventSubmitter = Optional.of(new EventSubmitter.Builder(metricContext, "org.apache.gobblin.service").build()); } else { this.eventSubmitter = Optional.absent(); } this.dagManagerMetrics = new DagManagerMetrics(); TimeUnit jobStartTimeUnit = TimeUnit.valueOf(ConfigUtils.getString(config, JOB_START_SLA_UNITS, ConfigurationKeys.FALLBACK_GOBBLIN_JOB_START_SLA_TIME_UNIT)); this.defaultJobStartSlaTimeMillis = jobStartTimeUnit.toMillis(ConfigUtils.getLong(config, JOB_START_SLA_TIME, ConfigurationKeys.FALLBACK_GOBBLIN_JOB_START_SLA_TIME)); this.jobStatusRetriever = jobStatusRetriever; this.flowStatusGenerator = flowStatusGenerator; this.specCompiler = GobblinConstructorUtils.invokeConstructor(SpecCompiler.class, ConfigUtils.getString(config, ServiceConfigKeys.GOBBLIN_SERVICE_FLOWCOMPILER_CLASS_KEY, ServiceConfigKeys.DEFAULT_GOBBLIN_SERVICE_FLOWCOMPILER_CLASS), config); this.isFlowConcurrencyEnabled = ConfigUtils.getBoolean(config, ServiceConfigKeys.FLOW_CONCURRENCY_ALLOWED, ServiceConfigKeys.DEFAULT_FLOW_CONCURRENCY_ALLOWED); this.quotaManager = GobblinConstructorUtils.invokeConstructor(UserQuotaManager.class, ConfigUtils.getString(config, ServiceConfigKeys.QUOTA_MANAGER_CLASS, ServiceConfigKeys.DEFAULT_QUOTA_MANAGER), config); this.flowCatalog = flowCatalog; this.flowCompilationValidationHelper = new FlowCompilationValidationHelper(sharedFlowMetricsSingleton, specCompiler, quotaManager, eventSubmitter, flowStatusGenerator, isFlowConcurrencyEnabled); TimeUnit timeUnit = TimeUnit.valueOf(ConfigUtils.getString(config, FAILED_DAG_RETENTION_TIME_UNIT, DEFAULT_FAILED_DAG_RETENTION_TIME_UNIT)); this.failedDagRetentionTime = timeUnit.toMillis(ConfigUtils.getLong(config, FAILED_DAG_RETENTION_TIME, DEFAULT_FAILED_DAG_RETENTION_TIME)); } DagStateStore createDagStateStore(Config config, Map<URI, TopologySpec> topologySpecMap) { try { Class<?> dagStateStoreClass = Class.forName(ConfigUtils.getString(config, DAG_STATESTORE_CLASS_KEY, FSDagStateStore.class.getName())); return (DagStateStore) GobblinConstructorUtils.invokeLongestConstructor(dagStateStoreClass, config, topologySpecMap); } catch (ReflectiveOperationException e) { throw new RuntimeException(e); } } // Initializes and returns an array of Queue of size numThreads private static LinkedBlockingDeque<?>[] initializeDagQueue(int numThreads) { LinkedBlockingDeque<?>[] queue = new LinkedBlockingDeque[numThreads]; for (int i=0; i< numThreads; i++) { queue[i] = new LinkedBlockingDeque<>(); } return queue; } @Inject public DagManager(Config config, JobStatusRetriever jobStatusRetriever, SharedFlowMetricsSingleton sharedFlowMetricsSingleton, FlowStatusGenerator flowStatusGenerator, FlowCatalog flowCatalog) { this(config, jobStatusRetriever, sharedFlowMetricsSingleton, flowStatusGenerator, flowCatalog, true); } /** Do Nothing on service startup. Scheduling of {@link DagManagerThread}s and loading of any {@link Dag}s is done * during leadership change. */ @Override protected void startUp() { //Do nothing. } /** * Method to submit a {@link Dag} to the {@link DagManager}. The {@link DagManager} optionally persists the * submitted dag to the {@link DagStateStore} and then adds the dag to a {@link BlockingQueue} to be picked up * by one of the {@link DagManagerThread}s. * @param dag {@link Dag} to be added * @param persist whether to persist the dag to the {@link DagStateStore} * @param setStatus if true, set all jobs in the dag to pending * Note this should only be called from the {@link Orchestrator} or {@link org.apache.gobblin.service.monitoring.DagActionStoreChangeMonitor} */ public synchronized void addDag(Dag<JobExecutionPlan> dag, boolean persist, boolean setStatus) throws IOException { // TODO: Used to track missing dag issue, remove later as needed log.info("Add dag (persist: {}, setStatus: {}): {}", persist, setStatus, dag); if (!isActive) { log.warn("Skipping add dag because this instance of DagManager is not active for dag: {}", dag); return; } if (persist) { //Persist the dag this.dagStateStore.writeCheckpoint(dag); } int queueId = DagManagerUtils.getDagQueueId(dag, this.numThreads); // Add the dag to the specific queue determined by flowExecutionId // Flow cancellation request has to be forwarded to the same DagManagerThread where the // flow create request was forwarded. This is because Azkaban Exec Id is stored in the DagNode of the // specific DagManagerThread queue if (!this.runQueue[queueId].offer(dag)) { throw new IOException("Could not add dag" + DagManagerUtils.generateDagId(dag) + "to queue"); } if (setStatus) { submitEventsAndSetStatus(dag); } } private void submitEventsAndSetStatus(Dag<JobExecutionPlan> dag) { if (this.eventSubmitter.isPresent()) { for (DagNode<JobExecutionPlan> dagNode : dag.getNodes()) { JobExecutionPlan jobExecutionPlan = DagManagerUtils.getJobExecutionPlan(dagNode); Map<String, String> jobMetadata = TimingEventUtils.getJobMetadata(Maps.newHashMap(), jobExecutionPlan); this.eventSubmitter.get().getTimingEvent(TimingEvent.LauncherTimings.JOB_PENDING).stop(jobMetadata); jobExecutionPlan.setExecutionStatus(PENDING); } } } /** * Method to submit a {@link URI} for cancellation requests to the {@link DagManager}. * The {@link DagManager} adds the dag to the {@link BlockingQueue} to be picked up by one of the {@link DagManagerThread}s. */ synchronized public void stopDag(URI uri) throws IOException { String flowGroup = FlowSpec.Utils.getFlowGroup(uri); String flowName = FlowSpec.Utils.getFlowName(uri); List<Long> flowExecutionIds = this.jobStatusRetriever.getLatestExecutionIdsForFlow(flowName, flowGroup, 10); log.info("Found {} flows to cancel.", flowExecutionIds.size()); for (long flowExecutionId : flowExecutionIds) { killFlow(flowGroup, flowName, flowExecutionId); } } /** * Add the specified flow to {@link DagManager#cancelQueue} */ private void killFlow(String flowGroup, String flowName, long flowExecutionId) throws IOException { int queueId = DagManagerUtils.getDagQueueId(flowExecutionId, this.numThreads); DagId dagId = DagManagerUtils.generateDagId(flowGroup, flowName, flowExecutionId); if (!this.cancelQueue[queueId].offer(dagId)) { throw new IOException("Could not add dag " + dagId + " to cancellation queue."); } } @Subscribe public void handleKillFlowEvent(KillFlowEvent killFlowEvent) { handleKillFlowRequest(killFlowEvent.getFlowGroup(), killFlowEvent.getFlowName(), killFlowEvent.getFlowExecutionId()); } // Method used to handle kill flow requests received from subscriber-event model or from direct invocation public void handleKillFlowRequest(String flowGroup, String flowName, long flowExecutionId) { if (isActive) { log.info("Received kill request for flow ({}, {}, {})", flowGroup, flowName, flowExecutionId); try { killFlow(flowGroup, flowName, flowExecutionId); } catch (IOException e) { log.warn("Failed to kill flow", e); } } } // Method used to handle resume flow requests received from subscriber-event model or from direct invocation public void handleResumeFlowRequest(String flowGroup, String flowName, long flowExecutionId) { if (isActive) { log.info("Received resume request for flow ({}, {}, {})", flowGroup, flowName, flowExecutionId); DagId dagId = DagManagerUtils.generateDagId(flowGroup, flowName, flowExecutionId); int queueId = DagManagerUtils.getDagQueueId(flowExecutionId, this.numThreads); if (!this.resumeQueue[queueId].offer(dagId)) { log.warn("Could not add dag " + dagId + " to resume queue"); } } } @Subscribe public void handleResumeFlowEvent(ResumeFlowEvent resumeFlowEvent) { handleResumeFlowRequest(resumeFlowEvent.getFlowGroup(), resumeFlowEvent.getFlowName(), resumeFlowEvent.getFlowExecutionId()); } public synchronized void setTopologySpecMap(Map<URI, TopologySpec> topologySpecMap) { this.topologySpecMap = topologySpecMap; } /** * When a {@link DagManager} becomes active, it loads the serialized representations of the currently running {@link Dag}s * from the checkpoint directory, deserializes the {@link Dag}s and adds them to a queue to be consumed by * the {@link DagManagerThread}s. * @param active a boolean to indicate if the {@link DagManager} is the leader. */ public synchronized void setActive(boolean active) { if (this.isActive == active) { log.info("DagManager already {}, skipping further actions.", (!active) ? "inactive" : "active"); return; } this.isActive = active; try { if (this.isActive) { log.info("Activating DagManager."); log.info("Scheduling {} DagManager threads", numThreads); //Initializing state store for persisting Dags. this.dagStateStore = createDagStateStore(config, topologySpecMap); DagStateStore failedDagStateStore = createDagStateStore(ConfigUtils.getConfigOrEmpty(config, FAILED_DAG_STATESTORE_PREFIX).withFallback(config), topologySpecMap); Set<String> failedDagIds = Collections.synchronizedSet(failedDagStateStore.getDagIds()); this.dagManagerMetrics.activate(); UserQuotaManager quotaManager = GobblinConstructorUtils.invokeConstructor(UserQuotaManager.class, ConfigUtils.getString(config, ServiceConfigKeys.QUOTA_MANAGER_CLASS, ServiceConfigKeys.DEFAULT_QUOTA_MANAGER), config); quotaManager.init(dagStateStore.getDags()); //On startup, the service creates DagManagerThreads that are scheduled at a fixed rate. this.dagManagerThreads = new DagManagerThread[numThreads]; for (int i = 0; i < numThreads; i++) { DagManagerThread dagManagerThread = new DagManagerThread(jobStatusRetriever, dagStateStore, failedDagStateStore, dagActionStore, runQueue[i], cancelQueue[i], resumeQueue[i], instrumentationEnabled, failedDagIds, this.dagManagerMetrics, this.defaultJobStartSlaTimeMillis, quotaManager, i); this.dagManagerThreads[i] = dagManagerThread; this.scheduledExecutorPool.scheduleAtFixedRate(dagManagerThread, 0, this.pollingInterval, TimeUnit.SECONDS); } FailedDagRetentionThread failedDagRetentionThread = new FailedDagRetentionThread(failedDagStateStore, failedDagIds, failedDagRetentionTime); this.scheduledExecutorPool.scheduleAtFixedRate(failedDagRetentionThread, 0, retentionPollingInterval, TimeUnit.MINUTES); loadDagFromDagStateStore(); this.houseKeepingThreadPool = Executors.newSingleThreadScheduledExecutor(); for (int delay = houseKeepingThreadInitialDelay; delay < MAX_HOUSEKEEPING_THREAD_DELAY; delay *= 2) { this.houseKeepingThreadPool.schedule(() -> { try { loadDagFromDagStateStore(); } catch (Exception e ) { log.error("failed to sync dag state store due to ", e); }}, delay, TimeUnit.MINUTES); } if (dagActionStore.isPresent()) { Collection<DagActionStore.DagAction> dagActions = dagActionStore.get().getDagActions(); for (DagActionStore.DagAction action : dagActions) { switch (action.getFlowActionType()) { case KILL: this.handleKillFlowEvent(new KillFlowEvent(action.getFlowGroup(), action.getFlowName(), Long.parseLong(action.getFlowExecutionId()))); break; case RESUME: this.handleResumeFlowEvent(new ResumeFlowEvent(action.getFlowGroup(), action.getFlowName(), Long.parseLong(action.getFlowExecutionId()))); break; case LAUNCH: this.handleLaunchFlowEvent(action); break; default: log.warn("Unsupported dagAction: " + action.getFlowActionType().toString()); } } } } else { //Mark the DagManager inactive. log.info("Inactivating the DagManager. Shutting down all DagManager threads"); this.scheduledExecutorPool.shutdown(); this.dagManagerMetrics.cleanup(); this.houseKeepingThreadPool.shutdown(); try { this.scheduledExecutorPool.awaitTermination(TERMINATION_TIMEOUT, TimeUnit.SECONDS); } catch (InterruptedException e) { log.error("Exception encountered when shutting down DagManager threads.", e); } } } catch (IOException e) { log.error("Exception encountered when activating the new DagManager", e); throw new RuntimeException(e); } } /** * Used by the DagManager to launch a new execution for a flow action event loaded from the DagActionStore upon * setting this instance of the DagManager to active. Because it may be a completely new DAG not contained in the * dagStore, we compile the flow to generate the dag before calling addDag(), handling any errors that may result in * the process. */ public void handleLaunchFlowEvent(DagActionStore.DagAction launchAction) { Preconditions.checkArgument(launchAction.getFlowActionType() == DagActionStore.FlowActionType.LAUNCH); log.info("Handle launch flow event for action {}", launchAction); FlowId flowId = launchAction.getFlowId(); try { URI flowUri = FlowSpec.Utils.createFlowSpecUri(flowId); FlowSpec spec = (FlowSpec) flowCatalog.getSpecs(flowUri); Optional<Dag<JobExecutionPlan>> optionalJobExecutionPlanDag = this.flowCompilationValidationHelper.createExecutionPlanIfValid(spec, Optional.absent()); if (optionalJobExecutionPlanDag.isPresent()) { addDag(optionalJobExecutionPlanDag.get(), true, true); } else { log.warn("Failed flow compilation of spec causing launch flow event to be skipped on startup. Flow {}", flowId); this.dagManagerMetrics.incrementFailedLaunchCount(); } // Upon handling the action, delete it so on leadership change this is not duplicated this.dagActionStore.get().deleteDagAction(launchAction); } catch (URISyntaxException e) { log.warn(String.format("Could not create URI object for flowId %s due to exception", flowId), e); this.dagManagerMetrics.incrementFailedLaunchCount(); } catch (SpecNotFoundException e) { log.warn(String.format("Spec not found for flowId %s due to exception", flowId), e); this.dagManagerMetrics.incrementFailedLaunchCount(); } catch (IOException e) { log.warn(String.format("Failed to add Job Execution Plan for flowId %s OR delete dag action from dagActionStore " + "(check stacktrace) due to exception", flowId), e); this.dagManagerMetrics.incrementFailedLaunchCount(); } catch (InterruptedException e) { log.warn(String.format("SpecCompiler failed to reach healthy state before compilation of flowId %s due to " + "exception", flowId), e); this.dagManagerMetrics.incrementFailedLaunchCount(); } } private void loadDagFromDagStateStore() throws IOException { List<Dag<JobExecutionPlan>> dags = dagStateStore.getDags(); log.info("Loading " + dags.size() + " dags from dag state store"); for (Dag<JobExecutionPlan> dag : dags) { if (this.isActive) { addDag(dag, false, false); } } } /** * Each {@link DagManagerThread} performs 2 actions when scheduled: * <ol> * <li> Dequeues any newly submitted {@link Dag}s from the Dag queue. All the {@link JobExecutionPlan}s which * are part of the dequed {@link Dag} will be managed this thread. </li> * <li> Polls the job status store for the current job statuses of all the running jobs it manages.</li> * </ol> */ public static class DagManagerThread implements Runnable { private final Map<DagNode<JobExecutionPlan>, Dag<JobExecutionPlan>> jobToDag = new HashMap<>(); private final Map<String, Dag<JobExecutionPlan>> dags = new HashMap<>(); private final Set<String> failedDagIds; private final Map<String, Dag<JobExecutionPlan>> resumingDags = new HashMap<>(); // dagToJobs holds a map of dagId to running jobs of that dag final Map<String, LinkedList<DagNode<JobExecutionPlan>>> dagToJobs = new HashMap<>(); final Map<String, Long> dagToSLA = new HashMap<>(); private final MetricContext metricContext; private final Set<String> dagIdstoClean = new HashSet<>(); private final Optional<EventSubmitter> eventSubmitter; private final Optional<Timer> jobStatusPolledTimer; private final AtomicLong orchestrationDelay = new AtomicLong(0); private final DagManagerMetrics dagManagerMetrics; private final UserQuotaManager quotaManager; private final JobStatusRetriever jobStatusRetriever; private final DagStateStore dagStateStore; private final DagStateStore failedDagStateStore; private final BlockingQueue<Dag<JobExecutionPlan>> queue; private final BlockingQueue<DagId> cancelQueue; private final BlockingQueue<DagId> resumeQueue; private final Long defaultJobStartSlaTimeMillis; private final Optional<DagActionStore> dagActionStore; private final Optional<Meter> dagManagerThreadHeartbeat; /** * Constructor. */ DagManagerThread(JobStatusRetriever jobStatusRetriever, DagStateStore dagStateStore, DagStateStore failedDagStateStore, Optional<DagActionStore> dagActionStore, BlockingQueue<Dag<JobExecutionPlan>> queue, BlockingQueue<DagId> cancelQueue, BlockingQueue<DagId> resumeQueue, boolean instrumentationEnabled, Set<String> failedDagIds, DagManagerMetrics dagManagerMetrics, Long defaultJobStartSla, UserQuotaManager quotaManager, int dagMangerThreadId) { this.jobStatusRetriever = jobStatusRetriever; this.dagStateStore = dagStateStore; this.failedDagStateStore = failedDagStateStore; this.failedDagIds = failedDagIds; this.queue = queue; this.cancelQueue = cancelQueue; this.resumeQueue = resumeQueue; this.dagManagerMetrics = dagManagerMetrics; this.defaultJobStartSlaTimeMillis = defaultJobStartSla; this.quotaManager = quotaManager; this.dagActionStore = dagActionStore; if (instrumentationEnabled) { this.metricContext = Instrumented.getMetricContext(ConfigUtils.configToState(ConfigFactory.empty()), getClass()); this.eventSubmitter = Optional.of(new EventSubmitter.Builder(this.metricContext, "org.apache.gobblin.service").build()); this.jobStatusPolledTimer = Optional.of(this.metricContext.timer(ServiceMetricNames.JOB_STATUS_POLLED_TIMER)); ContextAwareGauge<Long> orchestrationDelayMetric = metricContext.newContextAwareGauge(ServiceMetricNames.FLOW_ORCHESTRATION_DELAY, orchestrationDelay::get); this.metricContext.register(orchestrationDelayMetric); this.dagManagerThreadHeartbeat = Optional.of(this.metricContext.contextAwareMeter(String.format(DAG_MANAGER_HEARTBEAT, dagMangerThreadId))); } else { this.metricContext = null; this.eventSubmitter = Optional.absent(); this.jobStatusPolledTimer = Optional.absent(); this.dagManagerThreadHeartbeat = Optional.absent(); } } /** * Main body of the {@link DagManagerThread}. Deque the next item from the queue and poll job statuses of currently * running jobs. * Because this thread runs in a regular interval, we should avoid doing repetitive work inside it. */ @Override public void run() { try { DagId nextDagToCancel = cancelQueue.poll(); //Poll the cancelQueue for a new Dag to cancel. if (nextDagToCancel != null) { cancelDag(nextDagToCancel); } while (!queue.isEmpty()) { Dag<JobExecutionPlan> dag = queue.poll(); //Poll the queue for a new Dag to execute. if (dag != null) { if (dag.isEmpty()) { log.warn("Empty dag; ignoring the dag"); } //Initialize dag. initialize(dag); } else { log.warn("Null dag despite non-empty queue; ignoring the dag"); } } while (!resumeQueue.isEmpty()) { DagId dagId = resumeQueue.poll(); beginResumingDag(dagId); } finishResumingDags(); log.debug("Polling job statuses.."); //Poll and update the job statuses of running jobs. pollAndAdvanceDag(); log.debug("Poll done."); //Clean up any finished dags log.debug("Cleaning up finished dags.."); cleanUp(); log.debug("Clean up done"); Instrumented.markMeter(dagManagerThreadHeartbeat); } catch (Exception e) { log.error(String.format("Exception encountered in %s", getClass().getName()), e); } } private void removeDagActionFromStore(DagId dagId, DagActionStore.FlowActionType flowActionType) throws IOException { if (this.dagActionStore.isPresent()) { this.dagActionStore.get().deleteDagAction( new DagActionStore.DagAction(dagId.flowGroup, dagId.flowName, dagId.flowExecutionId, flowActionType)); } } /** * Begin resuming a dag by setting the status of both the dag and the failed/cancelled dag nodes to {@link ExecutionStatus#PENDING_RESUME}, * and also sending events so that this status will be reflected in the job status state store. */ private void beginResumingDag(DagId dagIdToResume) throws IOException { String dagId= dagIdToResume.toString(); if (!this.failedDagIds.contains(dagId)) { log.warn("No dag found with dagId " + dagId + ", so cannot resume flow"); removeDagActionFromStore(dagIdToResume, DagActionStore.FlowActionType.RESUME); return; } Dag<JobExecutionPlan> dag = this.failedDagStateStore.getDag(dagId); if (dag == null) { log.error("Dag " + dagId + " was found in memory but not found in failed dag state store"); removeDagActionFromStore(dagIdToResume, DagActionStore.FlowActionType.RESUME); return; } long flowResumeTime = System.currentTimeMillis(); // Set the flow and it's failed or cancelled nodes to PENDING_RESUME so that the flow will be resumed from the point before it failed DagManagerUtils.emitFlowEvent(this.eventSubmitter, dag, TimingEvent.FlowTimings.FLOW_PENDING_RESUME); for (DagNode<JobExecutionPlan> node : dag.getNodes()) { ExecutionStatus executionStatus = node.getValue().getExecutionStatus(); if (executionStatus.equals(FAILED) || executionStatus.equals(CANCELLED)) { node.getValue().setExecutionStatus(PENDING_RESUME); // reset currentAttempts because we do not want to count previous execution's attempts in deciding whether to retry a job node.getValue().setCurrentAttempts(0); DagManagerUtils.incrementJobGeneration(node); Map<String, String> jobMetadata = TimingEventUtils.getJobMetadata(Maps.newHashMap(), node.getValue()); this.eventSubmitter.get().getTimingEvent(TimingEvent.LauncherTimings.JOB_PENDING_RESUME).stop(jobMetadata); } // Set flowStartTime so that flow SLA will be based on current time instead of original flow node.getValue().setFlowStartTime(flowResumeTime); } this.resumingDags.put(dagId, dag); } /** * Finish resuming dags by first verifying the status is correct (flow should be {@link ExecutionStatus#PENDING_RESUME} * and jobs should not be {@link ExecutionStatus#FAILED} or {@link ExecutionStatus#CANCELLED}) and then calling * {@link #initialize}. This is separated from {@link #beginResumingDag} because it could take some time for the * job status state store to reflect the updated status. */ private void finishResumingDags() throws IOException { for (Map.Entry<String, Dag<JobExecutionPlan>> dag : this.resumingDags.entrySet()) { JobStatus flowStatus = pollFlowStatus(dag.getValue()); if (flowStatus == null || !flowStatus.getEventName().equals(PENDING_RESUME.name())) { continue; } boolean dagReady = true; for (DagNode<JobExecutionPlan> node : dag.getValue().getNodes()) { JobStatus jobStatus = pollJobStatus(node); if (jobStatus == null || jobStatus.getEventName().equals(FAILED.name()) || jobStatus.getEventName().equals(CANCELLED.name())) { dagReady = false; break; } } if (dagReady) { this.dagStateStore.writeCheckpoint(dag.getValue()); this.failedDagStateStore.cleanUp(dag.getValue()); removeDagActionFromStore(DagManagerUtils.generateDagId(dag.getValue()), DagActionStore.FlowActionType.RESUME); this.failedDagIds.remove(dag.getKey()); this.resumingDags.remove(dag.getKey()); initialize(dag.getValue()); } } } /** * Cancels the dag and sends a cancellation tracking event. * @param dagToCancel dag node to cancel * @throws ExecutionException executionException * @throws InterruptedException interruptedException */ private void cancelDag(DagId dagId) throws ExecutionException, InterruptedException, IOException { String dagToCancel = dagId.toString(); log.info("Cancel flow with DagId {}", dagToCancel); if (this.dagToJobs.containsKey(dagToCancel)) { List<DagNode<JobExecutionPlan>> dagNodesToCancel = this.dagToJobs.get(dagToCancel); log.info("Found {} DagNodes to cancel.", dagNodesToCancel.size()); for (DagNode<JobExecutionPlan> dagNodeToCancel : dagNodesToCancel) { cancelDagNode(dagNodeToCancel); } this.dags.get(dagToCancel).setFlowEvent(TimingEvent.FlowTimings.FLOW_CANCELLED); this.dags.get(dagToCancel).setMessage("Flow killed by request"); } else { log.warn("Did not find Dag with id {}, it might be already cancelled/finished.", dagToCancel); } // Called after a KILL request is received removeDagActionFromStore(dagId, DagActionStore.FlowActionType.KILL); } private void cancelDagNode(DagNode<JobExecutionPlan> dagNodeToCancel) throws ExecutionException, InterruptedException { Properties props = new Properties(); if (dagNodeToCancel.getValue().getJobFuture().isPresent()) { Future future = dagNodeToCancel.getValue().getJobFuture().get(); String serializedFuture = DagManagerUtils.getSpecProducer(dagNodeToCancel).serializeAddSpecResponse(future); props.put(ConfigurationKeys.SPEC_PRODUCER_SERIALIZED_FUTURE, serializedFuture); sendCancellationEvent(dagNodeToCancel.getValue()); } if (dagNodeToCancel.getValue().getJobSpec().getConfig().hasPath(ConfigurationKeys.FLOW_EXECUTION_ID_KEY)) { props.setProperty(ConfigurationKeys.FLOW_EXECUTION_ID_KEY, dagNodeToCancel.getValue().getJobSpec().getConfig().getString(ConfigurationKeys.FLOW_EXECUTION_ID_KEY)); } DagManagerUtils.getSpecProducer(dagNodeToCancel).cancelJob(dagNodeToCancel.getValue().getJobSpec().getUri(), props); } private void sendCancellationEvent(JobExecutionPlan jobExecutionPlan) { if (this.eventSubmitter.isPresent()) { Map<String, String> jobMetadata = TimingEventUtils.getJobMetadata(Maps.newHashMap(), jobExecutionPlan); this.eventSubmitter.get().getTimingEvent(TimingEvent.LauncherTimings.JOB_CANCEL).stop(jobMetadata); jobExecutionPlan.setExecutionStatus(CANCELLED); } } /** * This method determines the next set of jobs to execute from the dag and submits them for execution. * This method updates internal data structures tracking currently running Dags and jobs. */ private void initialize(Dag<JobExecutionPlan> dag) throws IOException { //Add Dag to the map of running dags String dagId = DagManagerUtils.generateDagId(dag).toString(); log.info("Initializing Dag {}", DagManagerUtils.getFullyQualifiedDagName(dag)); if (this.dags.containsKey(dagId)) { log.warn("Already tracking a dag with dagId {}, skipping.", dagId); return; } this.dags.put(dagId, dag); log.debug("Dag {} - determining if any jobs are already running.", DagManagerUtils.getFullyQualifiedDagName(dag)); //A flag to indicate if the flow is already running. boolean isDagRunning = false; //Are there any jobs already in the running state? This check is for Dags already running //before a leadership change occurs. for (DagNode<JobExecutionPlan> dagNode : dag.getNodes()) { if (DagManagerUtils.getExecutionStatus(dagNode) == RUNNING) { addJobState(dagId, dagNode); //Update the running jobs counter. dagManagerMetrics.incrementRunningJobMetrics(dagNode); isDagRunning = true; } } FlowId flowId = DagManagerUtils.getFlowId(dag); this.dagManagerMetrics.registerFlowMetric(flowId, dag); log.debug("Dag {} submitting jobs ready for execution.", DagManagerUtils.getFullyQualifiedDagName(dag)); //Determine the next set of jobs to run and submit them for execution Map<String, Set<DagNode<JobExecutionPlan>>> nextSubmitted = submitNext(dagId); for (DagNode<JobExecutionPlan> dagNode: nextSubmitted.get(dagId)) { addJobState(dagId, dagNode); } // Set flow status to running DagManagerUtils.emitFlowEvent(this.eventSubmitter, dag, TimingEvent.FlowTimings.FLOW_RUNNING); dagManagerMetrics.conditionallyMarkFlowAsState(flowId, FlowState.RUNNING); // Report the orchestration delay the first time the Dag is initialized. Orchestration delay is defined as // the time difference between the instant when a flow first transitions to the running state and the instant // when the flow is submitted to Gobblin service. if (!isDagRunning) { this.orchestrationDelay.set(System.currentTimeMillis() - DagManagerUtils.getFlowExecId(dag)); } log.info("Dag {} Initialization complete.", DagManagerUtils.getFullyQualifiedDagName(dag)); } /** * Proceed the execution of each dag node based on job status. */ private void pollAndAdvanceDag() throws IOException, ExecutionException, InterruptedException { Map<String, Set<DagNode<JobExecutionPlan>>> nextSubmitted = Maps.newHashMap(); List<DagNode<JobExecutionPlan>> nodesToCleanUp = Lists.newArrayList(); for (DagNode<JobExecutionPlan> node : this.jobToDag.keySet()) { try { boolean slaKilled = slaKillIfNeeded(node); JobStatus jobStatus = pollJobStatus(node); boolean killOrphanFlow = killJobIfOrphaned(node, jobStatus); ExecutionStatus status = getJobExecutionStatus(slaKilled, killOrphanFlow, jobStatus); JobExecutionPlan jobExecutionPlan = DagManagerUtils.getJobExecutionPlan(node); switch (status) { case COMPLETE: jobExecutionPlan.setExecutionStatus(COMPLETE); nextSubmitted.putAll(onJobFinish(node)); nodesToCleanUp.add(node); break; case FAILED: jobExecutionPlan.setExecutionStatus(FAILED); nextSubmitted.putAll(onJobFinish(node)); nodesToCleanUp.add(node); break; case CANCELLED: jobExecutionPlan.setExecutionStatus(CANCELLED); nextSubmitted.putAll(onJobFinish(node)); nodesToCleanUp.add(node); break; case PENDING: jobExecutionPlan.setExecutionStatus(PENDING); break; case PENDING_RETRY: jobExecutionPlan.setExecutionStatus(PENDING_RETRY); break; default: jobExecutionPlan.setExecutionStatus(RUNNING); break; } if (jobStatus != null && jobStatus.isShouldRetry()) { log.info("Retrying job: {}, current attempts: {}, max attempts: {}", DagManagerUtils.getFullyQualifiedJobName(node), jobStatus.getCurrentAttempts(), jobStatus.getMaxAttempts()); this.jobToDag.get(node).setFlowEvent(null); submitJob(node); } } catch (Exception e) { // Error occurred while processing dag, continue processing other dags assigned to this thread log.error(String.format("Exception caught in DagManager while processing dag %s due to ", DagManagerUtils.getFullyQualifiedDagName(node)), e); } } for (Map.Entry<String, Set<DagNode<JobExecutionPlan>>> entry: nextSubmitted.entrySet()) { String dagId = entry.getKey(); Set<DagNode<JobExecutionPlan>> dagNodes = entry.getValue(); for (DagNode<JobExecutionPlan> dagNode: dagNodes) { addJobState(dagId, dagNode); } } for (DagNode<JobExecutionPlan> dagNode: nodesToCleanUp) { String dagId = DagManagerUtils.generateDagId(dagNode).toString(); deleteJobState(dagId, dagNode); } } /** * Cancel the job if the job has been "orphaned". A job is orphaned if has been in ORCHESTRATED * {@link ExecutionStatus} for some specific amount of time. * @param node {@link DagNode} representing the job * @param jobStatus current {@link JobStatus} of the job * @return true if the total time that the job remains in the ORCHESTRATED state exceeds * {@value ConfigurationKeys#GOBBLIN_JOB_START_SLA_TIME}. */ private boolean killJobIfOrphaned(DagNode<JobExecutionPlan> node, JobStatus jobStatus) throws ExecutionException, InterruptedException { if (jobStatus == null) { return false; } ExecutionStatus executionStatus = valueOf(jobStatus.getEventName()); long timeOutForJobStart = DagManagerUtils.getJobStartSla(node, this.defaultJobStartSlaTimeMillis); long jobOrchestratedTime = jobStatus.getOrchestratedTime(); if (executionStatus == ORCHESTRATED && System.currentTimeMillis() - jobOrchestratedTime > timeOutForJobStart) { log.info("Job {} of flow {} exceeded the job start SLA of {} ms. Killing the job now...", DagManagerUtils.getJobName(node), DagManagerUtils.getFullyQualifiedDagName(node), timeOutForJobStart); dagManagerMetrics.incrementCountsStartSlaExceeded(node); cancelDagNode(node); String dagId = DagManagerUtils.generateDagId(node).toString(); this.dags.get(dagId).setFlowEvent(TimingEvent.FlowTimings.FLOW_START_DEADLINE_EXCEEDED); this.dags.get(dagId).setMessage("Flow killed because no update received for " + timeOutForJobStart + " ms after orchestration"); return true; } else { return false; } } private ExecutionStatus getJobExecutionStatus(boolean slaKilled, boolean killOrphanFlow, JobStatus jobStatus) { if (slaKilled || killOrphanFlow) { return CANCELLED; } else { if (jobStatus == null) { return PENDING; } else { return valueOf(jobStatus.getEventName()); } } } /** * Check if the SLA is configured for the flow this job belongs to. * If it is, this method will try to cancel the job when SLA is reached. * * @param node dag node of the job * @return true if the job is killed because it reached sla * @throws ExecutionException exception * @throws InterruptedException exception */ private boolean slaKillIfNeeded(DagNode<JobExecutionPlan> node) throws ExecutionException, InterruptedException { long flowStartTime = DagManagerUtils.getFlowStartTime(node); long currentTime = System.currentTimeMillis(); String dagId = DagManagerUtils.generateDagId(node).toString(); long flowSla; if (dagToSLA.containsKey(dagId)) { flowSla = dagToSLA.get(dagId); } else { try { flowSla = DagManagerUtils.getFlowSLA(node); } catch (ConfigException e) { log.warn("Flow SLA for flowGroup: {}, flowName: {} is given in invalid format, using default SLA of {}", node.getValue().getJobSpec().getConfig().getString(ConfigurationKeys.FLOW_GROUP_KEY), node.getValue().getJobSpec().getConfig().getString(ConfigurationKeys.FLOW_NAME_KEY), DagManagerUtils.DEFAULT_FLOW_SLA_MILLIS); flowSla = DagManagerUtils.DEFAULT_FLOW_SLA_MILLIS; } dagToSLA.put(dagId, flowSla); } if (currentTime > flowStartTime + flowSla) { log.info("Flow {} exceeded the SLA of {} ms. Killing the job {} now...", node.getValue().getJobSpec().getConfig().getString(ConfigurationKeys.FLOW_NAME_KEY), flowSla, node.getValue().getJobSpec().getConfig().getString(ConfigurationKeys.JOB_NAME_KEY)); dagManagerMetrics.incrementExecutorSlaExceeded(node); cancelDagNode(node); this.dags.get(dagId).setFlowEvent(TimingEvent.FlowTimings.FLOW_RUN_DEADLINE_EXCEEDED); this.dags.get(dagId).setMessage("Flow killed due to exceeding SLA of " + flowSla + " ms"); return true; } return false; } /** * Retrieve the {@link JobStatus} from the {@link JobExecutionPlan}. */ private JobStatus pollJobStatus(DagNode<JobExecutionPlan> dagNode) { Config jobConfig = dagNode.getValue().getJobSpec().getConfig(); String flowGroup = jobConfig.getString(ConfigurationKeys.FLOW_GROUP_KEY); String flowName = jobConfig.getString(ConfigurationKeys.FLOW_NAME_KEY); long flowExecutionId = jobConfig.getLong(ConfigurationKeys.FLOW_EXECUTION_ID_KEY); String jobGroup = jobConfig.getString(ConfigurationKeys.JOB_GROUP_KEY); String jobName = jobConfig.getString(ConfigurationKeys.JOB_NAME_KEY); return pollStatus(flowGroup, flowName, flowExecutionId, jobGroup, jobName); } /** * Retrieve the flow's {@link JobStatus} (i.e. job status with {@link JobStatusRetriever#NA_KEY} as job name/group) from a dag */ private JobStatus pollFlowStatus(Dag<JobExecutionPlan> dag) { if (dag == null || dag.isEmpty()) { return null; } Config jobConfig = dag.getNodes().get(0).getValue().getJobSpec().getConfig(); String flowGroup = jobConfig.getString(ConfigurationKeys.FLOW_GROUP_KEY); String flowName = jobConfig.getString(ConfigurationKeys.FLOW_NAME_KEY); long flowExecutionId = jobConfig.getLong(ConfigurationKeys.FLOW_EXECUTION_ID_KEY); return pollStatus(flowGroup, flowName, flowExecutionId, JobStatusRetriever.NA_KEY, JobStatusRetriever.NA_KEY); } private JobStatus pollStatus(String flowGroup, String flowName, long flowExecutionId, String jobGroup, String jobName) { long pollStartTime = System.nanoTime(); Iterator<JobStatus> jobStatusIterator = this.jobStatusRetriever.getJobStatusesForFlowExecution(flowName, flowGroup, flowExecutionId, jobName, jobGroup); Instrumented.updateTimer(this.jobStatusPolledTimer, System.nanoTime() - pollStartTime, TimeUnit.NANOSECONDS); if (jobStatusIterator.hasNext()) { return jobStatusIterator.next(); } else { return null; } } /** * Submit next set of Dag nodes in the Dag identified by the provided dagId * @param dagId The dagId that should be processed. * @return * @throws IOException */ synchronized Map<String, Set<DagNode<JobExecutionPlan>>> submitNext(String dagId) throws IOException { Dag<JobExecutionPlan> dag = this.dags.get(dagId); Set<DagNode<JobExecutionPlan>> nextNodes = DagManagerUtils.getNext(dag); List<String> nextJobNames = new ArrayList<>(); //Submit jobs from the dag ready for execution. for (DagNode<JobExecutionPlan> dagNode : nextNodes) { submitJob(dagNode); nextJobNames.add(DagManagerUtils.getJobName(dagNode)); } log.info("Submitting next nodes for dagId {}, where next jobs to be submitted are {}", dagId, nextJobNames); //Checkpoint the dag state this.dagStateStore.writeCheckpoint(dag); Map<String, Set<DagNode<JobExecutionPlan>>> dagIdToNextJobs = Maps.newHashMap(); dagIdToNextJobs.put(dagId, nextNodes); return dagIdToNextJobs; } /** * Submits a {@link JobSpec} to a {@link org.apache.gobblin.runtime.api.SpecExecutor}. */ private void submitJob(DagNode<JobExecutionPlan> dagNode) { DagManagerUtils.incrementJobAttempt(dagNode); JobExecutionPlan jobExecutionPlan = DagManagerUtils.getJobExecutionPlan(dagNode); jobExecutionPlan.setExecutionStatus(RUNNING); JobSpec jobSpec = DagManagerUtils.getJobSpec(dagNode); Map<String, String> jobMetadata = TimingEventUtils.getJobMetadata(Maps.newHashMap(), jobExecutionPlan); String specExecutorUri = DagManagerUtils.getSpecExecutorUri(dagNode); // Run this spec on selected executor SpecProducer<Spec> producer; try { quotaManager.checkQuota(Collections.singleton(dagNode)); producer = DagManagerUtils.getSpecProducer(dagNode); TimingEvent jobOrchestrationTimer = this.eventSubmitter.isPresent() ? this.eventSubmitter.get(). getTimingEvent(TimingEvent.LauncherTimings.JOB_ORCHESTRATED) : null; // Increment job count before submitting the job onto the spec producer, in case that throws an exception. // By this point the quota is allocated, so it's imperative to increment as missing would introduce the potential to decrement below zero upon quota release. // Quota release is guaranteed, despite failure, because exception handling within would mark the job FAILED. // When the ensuing kafka message spurs DagManager processing, the quota is released and the counts decremented // Ensure that we do not double increment for flows that are retried if (dagNode.getValue().getCurrentAttempts() == 1) { dagManagerMetrics.incrementRunningJobMetrics(dagNode); } // Submit the job to the SpecProducer, which in turn performs the actual job submission to the SpecExecutor instance. // The SpecProducer implementations submit the job to the underlying executor and return when the submission is complete, // either successfully or unsuccessfully. To catch any exceptions in the job submission, the DagManagerThread // blocks (by calling Future#get()) until the submission is completed. Future<?> addSpecFuture = producer.addSpec(jobSpec); dagNode.getValue().setJobFuture(Optional.of(addSpecFuture)); //Persist the dag this.dagStateStore.writeCheckpoint(this.dags.get(DagManagerUtils.generateDagId(dagNode).toString())); addSpecFuture.get(); jobMetadata.put(TimingEvent.METADATA_MESSAGE, producer.getExecutionLink(addSpecFuture, specExecutorUri)); // Add serialized job properties as part of the orchestrated job event metadata jobMetadata.put(JobExecutionPlan.JOB_PROPS_KEY, dagNode.getValue().toString()); if (jobOrchestrationTimer != null) { jobOrchestrationTimer.stop(jobMetadata); } log.info("Orchestrated job: {} on Executor: {}", DagManagerUtils.getFullyQualifiedJobName(dagNode), specExecutorUri); this.dagManagerMetrics.incrementJobsSentToExecutor(dagNode); } catch (Exception e) { TimingEvent jobFailedTimer = this.eventSubmitter.isPresent() ? this.eventSubmitter.get(). getTimingEvent(TimingEvent.LauncherTimings.JOB_FAILED) : null; String message = "Cannot submit job " + DagManagerUtils.getFullyQualifiedJobName(dagNode) + " on executor " + specExecutorUri; log.error(message, e); jobMetadata.put(TimingEvent.METADATA_MESSAGE, message + " due to " + e.getMessage()); if (jobFailedTimer != null) { jobFailedTimer.stop(jobMetadata); } } } /** * Method that defines the actions to be performed when a job finishes either successfully or with failure. * This method updates the state of the dag and performs clean up actions as necessary. */ private Map<String, Set<DagNode<JobExecutionPlan>>> onJobFinish(DagNode<JobExecutionPlan> dagNode) throws IOException { Dag<JobExecutionPlan> dag = this.jobToDag.get(dagNode); String dagId = DagManagerUtils.generateDagId(dag).toString(); String jobName = DagManagerUtils.getFullyQualifiedJobName(dagNode); ExecutionStatus jobStatus = DagManagerUtils.getExecutionStatus(dagNode); log.info("Job {} of Dag {} has finished with status {}", jobName, dagId, jobStatus.name()); // Only decrement counters and quota for jobs that actually ran on the executor, not from a GaaS side failure/skip event if (quotaManager.releaseQuota(dagNode)) { dagManagerMetrics.decrementRunningJobMetrics(dagNode); } switch (jobStatus) { case FAILED: dag.setMessage("Flow failed because job " + jobName + " failed"); dag.setFlowEvent(TimingEvent.FlowTimings.FLOW_FAILED); dagManagerMetrics.incrementExecutorFailed(dagNode); return Maps.newHashMap(); case CANCELLED: dag.setFlowEvent(TimingEvent.FlowTimings.FLOW_CANCELLED); return Maps.newHashMap(); case COMPLETE: dagManagerMetrics.incrementExecutorSuccess(dagNode); return submitNext(dagId); default: log.warn("It should not reach here. Job status is unexpected."); return Maps.newHashMap(); } } private void deleteJobState(String dagId, DagNode<JobExecutionPlan> dagNode) { this.jobToDag.remove(dagNode); this.dagToJobs.get(dagId).remove(dagNode); this.dagToSLA.remove(dagId); } private void addJobState(String dagId, DagNode<JobExecutionPlan> dagNode) { Dag<JobExecutionPlan> dag = this.dags.get(dagId); this.jobToDag.put(dagNode, dag); if (this.dagToJobs.containsKey(dagId)) { this.dagToJobs.get(dagId).add(dagNode); } else { LinkedList<DagNode<JobExecutionPlan>> dagNodeList = Lists.newLinkedList(); dagNodeList.add(dagNode); this.dagToJobs.put(dagId, dagNodeList); } } private boolean hasRunningJobs(String dagId) { List<DagNode<JobExecutionPlan>> dagNodes = this.dagToJobs.get(dagId); return dagNodes != null && !dagNodes.isEmpty(); } /** * Perform clean up. Remove a dag from the dagstore if the dag is complete and update internal state. */ private void cleanUp() { // Approximate the time when the flow events are emitted to account for delay when the flow event is received by the job monitor long cleanUpProcessingTime = System.currentTimeMillis(); // Remove dags that are finished and emit their appropriate metrics for (Map.Entry<String, Dag<JobExecutionPlan>> dagIdKeyPair : this.dags.entrySet()) { String dagId = dagIdKeyPair.getKey(); // On service restart, we repopulate the dags that are waiting to be cleaned up if (dagIdstoClean.contains(dagId)) { continue; } Dag<JobExecutionPlan> dag = dagIdKeyPair.getValue(); if ((TimingEvent.FlowTimings.FLOW_FAILED.equals(dag.getFlowEvent()) || TimingEvent.FlowTimings.FLOW_CANCELLED.equals(dag.getFlowEvent())) && DagManagerUtils.getFailureOption(dag) == FailureOption.FINISH_RUNNING) { //Skip monitoring of any other jobs of the failed dag. LinkedList<DagNode<JobExecutionPlan>> dagNodeList = this.dagToJobs.get(dagId); while (!dagNodeList.isEmpty()) { DagNode<JobExecutionPlan> dagNode = dagNodeList.poll(); deleteJobState(dagId, dagNode); } } if (!hasRunningJobs(dagId)) { // Collect all the dagIds that are finished this.dagIdstoClean.add(dagId); if (dag.getFlowEvent() == null) { // If the dag flow event is not set, then it is successful dag.setFlowEvent(TimingEvent.FlowTimings.FLOW_SUCCEEDED); } else { addFailedDag(dagId, dag); } // send an event before cleaning up dag DagManagerUtils.emitFlowEvent(this.eventSubmitter, this.dags.get(dagId), dag.getFlowEvent()); dag.setEventEmittedTimeMillis(cleanUpProcessingTime); } } // Only clean up dags after the job status monitor processed the flow event for (Iterator<String> dagIdIterator = this.dagIdstoClean.iterator(); dagIdIterator.hasNext();) { String dagId = dagIdIterator.next(); Dag<JobExecutionPlan> dag = this.dags.get(dagId); JobStatus flowStatus = pollFlowStatus(dag); if (flowStatus != null && FlowStatusGenerator.FINISHED_STATUSES.contains(flowStatus.getEventName())) { FlowId flowId = DagManagerUtils.getFlowId(dag); switch(dag.getFlowEvent()) { case TimingEvent.FlowTimings.FLOW_SUCCEEDED: this.dagManagerMetrics.emitFlowSuccessMetrics(flowId); this.dagManagerMetrics.conditionallyMarkFlowAsState(flowId, FlowState.SUCCESSFUL); break; case TimingEvent.FlowTimings.FLOW_FAILED: this.dagManagerMetrics.emitFlowFailedMetrics(flowId); this.dagManagerMetrics.conditionallyMarkFlowAsState(flowId, FlowState.FAILED); break; case TimingEvent.FlowTimings.FLOW_CANCELLED: this.dagManagerMetrics.emitFlowSlaExceededMetrics(flowId); this.dagManagerMetrics.conditionallyMarkFlowAsState(flowId, FlowState.FAILED); break; default: log.warn("Unexpected flow event {} for dag {}", dag.getFlowEvent(), dagId); } log.info("Dag {} has finished with status {}; Cleaning up dag from the state store.", dagId, dag.getFlowEvent()); cleanUpDag(dagId); dagIdIterator.remove(); } else if (cleanUpProcessingTime > dag.getEventEmittedTimeMillis() + DAG_FLOW_STATUS_TOLERANCE_TIME_MILLIS) { // Re-emit the flow event if the flow status has not been processed within the DagFlowStatusTolerance time DagManagerUtils.emitFlowEvent(this.eventSubmitter, dag, dag.getFlowEvent()); } else { log.info("Waiting for flow event {} to be emitted before cleaning up dag {}", dag.getFlowEvent(), dagId); } } } /** * Add a dag to failed dag state store */ private synchronized void addFailedDag(String dagId, Dag<JobExecutionPlan> dag) { try { log.info("Adding dag " + dagId + " to failed dag state store"); this.failedDagStateStore.writeCheckpoint(this.dags.get(dagId)); } catch (IOException e) { log.error("Failed to add dag " + dagId + " to failed dag state store", e); } this.failedDagIds.add(dagId); } /** * Note that removal of a {@link Dag} entry in {@link #dags} needs to be happen after {@link #cleanUp()} * since the real {@link Dag} object is required for {@link #cleanUp()}, * and cleaning of all relevant states need to be atomic * @param dagId */ private synchronized void cleanUpDag(String dagId) { log.info("Cleaning up dagId {}", dagId); // clears flow event after cancelled job to allow resume event status to be set this.dags.get(dagId).setFlowEvent(null); try { this.dagStateStore.cleanUp(dags.get(dagId)); } catch (IOException ioe) { log.error(String.format("Failed to clean %s from backStore due to:", dagId), ioe); } this.dags.remove(dagId); this.dagToJobs.remove(dagId); } } public enum FlowState { FAILED(-1), RUNNING(0), SUCCESSFUL(1); public int value; FlowState(int value) { this.value = value; } } /** * Thread that runs retention on failed dags based on their original start time (which is the flow execution ID). */ public static class FailedDagRetentionThread implements Runnable { private final DagStateStore failedDagStateStore; private final Set<String> failedDagIds; private final long failedDagRetentionTime; FailedDagRetentionThread(DagStateStore failedDagStateStore, Set<String> failedDagIds, long failedDagRetentionTime) { this.failedDagStateStore = failedDagStateStore; this.failedDagIds = failedDagIds; this.failedDagRetentionTime = failedDagRetentionTime; } @Override public void run() { try { log.info("Cleaning failed dag state store"); long startTime = System.currentTimeMillis(); int numCleaned = 0; Set<String> failedDagIdsCopy = new HashSet<>(this.failedDagIds); for (String dagId : failedDagIdsCopy) { if (this.failedDagRetentionTime > 0L && startTime > DagManagerUtils.getFlowExecId(dagId) + this.failedDagRetentionTime) { this.failedDagStateStore.cleanUp(dagId); this.failedDagIds.remove(dagId); numCleaned++; } } log.info("Cleaned " + numCleaned + " dags from the failed dag state store"); } catch (Exception e) { log.error("Failed to run retention on failed dag state store", e); } } } /** Stop the service. */ @Override protected void shutDown() throws Exception { this.scheduledExecutorPool.shutdown(); this.scheduledExecutorPool.awaitTermination(TERMINATION_TIMEOUT, TimeUnit.SECONDS); } }
3,900
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/orchestration/DagManagerUtils.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.orchestration; import java.io.IOException; import java.util.Collections; import java.util.HashSet; import java.util.LinkedList; import java.util.List; import java.util.Map; import java.util.Properties; import java.util.Set; import java.util.concurrent.ExecutionException; import java.util.concurrent.TimeUnit; import java.util.stream.Collectors; import com.google.common.base.Optional; import com.google.common.collect.ImmutableMap; import com.google.common.collect.Lists; import com.typesafe.config.Config; import com.typesafe.config.ConfigFactory; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.metrics.event.EventSubmitter; import org.apache.gobblin.metrics.event.TimingEvent; import org.apache.gobblin.runtime.api.JobSpec; import org.apache.gobblin.runtime.api.Spec; import org.apache.gobblin.runtime.api.SpecProducer; import org.apache.gobblin.service.ExecutionStatus; import org.apache.gobblin.service.FlowId; import org.apache.gobblin.service.RequesterService; import org.apache.gobblin.service.ServiceRequester; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.flowgraph.Dag.DagNode; import org.apache.gobblin.service.modules.orchestration.DagManager.FailureOption; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.util.ConfigUtils; public class DagManagerUtils { static long DEFAULT_FLOW_SLA_MILLIS = TimeUnit.HOURS.toMillis(24); static String QUOTA_KEY_SEPERATOR = ","; static FlowId getFlowId(Dag<JobExecutionPlan> dag) { return getFlowId(dag.getStartNodes().get(0)); } static FlowId getFlowId(DagNode<JobExecutionPlan> dagNode) { Config jobConfig = dagNode.getValue().getJobSpec().getConfig(); String flowGroup = jobConfig.getString(ConfigurationKeys.FLOW_GROUP_KEY); String flowName = jobConfig.getString(ConfigurationKeys.FLOW_NAME_KEY); return new FlowId().setFlowGroup(flowGroup).setFlowName(flowName); } static long getFlowExecId(Dag<JobExecutionPlan> dag) { return getFlowExecId(dag.getStartNodes().get(0)); } static long getFlowExecId(DagNode<JobExecutionPlan> dagNode) { return getFlowExecId(dagNode.getValue().getJobSpec()); } static long getFlowExecId(JobSpec jobSpec) { return jobSpec.getConfig().getLong(ConfigurationKeys.FLOW_EXECUTION_ID_KEY); } static long getFlowExecId(String dagId) { return Long.parseLong(dagId.substring(dagId.lastIndexOf('_') + 1)); } /** * Generate a dagId object from the given {@link Dag} instance. * @param dag instance of a {@link Dag}. * @return a DagId object associated corresponding to the {@link Dag} instance. */ public static DagManager.DagId generateDagId(Dag<JobExecutionPlan> dag) { return generateDagId(dag.getStartNodes().get(0).getValue().getJobSpec().getConfig()); } private static DagManager.DagId generateDagId(Config jobConfig) { String flowGroup = jobConfig.getString(ConfigurationKeys.FLOW_GROUP_KEY); String flowName = jobConfig.getString(ConfigurationKeys.FLOW_NAME_KEY); long flowExecutionId = jobConfig.getLong(ConfigurationKeys.FLOW_EXECUTION_ID_KEY); return new DagManager.DagId(flowGroup, flowName, String.valueOf(flowExecutionId)); } static DagManager.DagId generateDagId(Dag.DagNode<JobExecutionPlan> dagNode) { return generateDagId(dagNode.getValue().getJobSpec().getConfig()); } static DagManager.DagId generateDagId(String flowGroup, String flowName, long flowExecutionId) { return generateDagId(flowGroup, flowName, String.valueOf(flowExecutionId)); } static DagManager.DagId generateDagId(String flowGroup, String flowName, String flowExecutionId) { return new DagManager.DagId(flowGroup, flowName, flowExecutionId); } /** * Returns a fully-qualified {@link Dag} name that includes: (flowGroup, flowName, flowExecutionId). * @param dag * @return fully qualified name of the underlying {@link Dag}. */ static String getFullyQualifiedDagName(Dag<JobExecutionPlan> dag) { FlowId flowid = getFlowId(dag); long flowExecutionId = getFlowExecId(dag); return "(flowGroup: " + flowid.getFlowGroup() + ", flowName: " + flowid.getFlowName() + ", flowExecutionId: " + flowExecutionId + ")"; } /** * Returns a fully-qualified {@link Dag} name that includes: (flowGroup, flowName, flowExecutionId). * @param dagNode * @return fully qualified name of the underlying {@link Dag}. */ static String getFullyQualifiedDagName(DagNode<JobExecutionPlan> dagNode) { FlowId flowid = getFlowId(dagNode); long flowExecutionId = getFlowExecId(dagNode); return "(flowGroup: " + flowid.getFlowGroup() + ", flowName: " + flowid.getFlowName() + ", flowExecutionId: " + flowExecutionId + ")"; } static String getJobName(DagNode<JobExecutionPlan> dagNode) { return dagNode.getValue().getJobSpec().getConfig().getString(ConfigurationKeys.JOB_NAME_KEY); } /** * Returns a fully-qualified job name that includes: (flowGroup, flowName, flowExecutionId, jobName). * @param dagNode * @return a fully qualified name of the underlying job. */ static String getFullyQualifiedJobName(DagNode<JobExecutionPlan> dagNode) { Config jobConfig = dagNode.getValue().getJobSpec().getConfig(); String flowGroup = ConfigUtils.getString(jobConfig, ConfigurationKeys.FLOW_GROUP_KEY, ""); String flowName = ConfigUtils.getString(jobConfig, ConfigurationKeys.FLOW_NAME_KEY, ""); Long flowExecutionId = ConfigUtils.getLong(jobConfig, ConfigurationKeys.FLOW_EXECUTION_ID_KEY, 0L); String jobName = ConfigUtils.getString(jobConfig, ConfigurationKeys.JOB_NAME_KEY, ""); return "(flowGroup: " + flowGroup + ", flowName: " + flowName + ", flowExecutionId: " + flowExecutionId + ", jobName: " + jobName + ")"; } static JobExecutionPlan getJobExecutionPlan(DagNode<JobExecutionPlan> dagNode) { return dagNode.getValue(); } public static JobSpec getJobSpec(DagNode<JobExecutionPlan> dagNode) { JobSpec jobSpec = dagNode.getValue().getJobSpec(); Map<String, String> configWithCurrentAttempts = ImmutableMap.of(ConfigurationKeys.JOB_CURRENT_ATTEMPTS, String.valueOf(dagNode.getValue().getCurrentAttempts()), ConfigurationKeys.JOB_CURRENT_GENERATION, String.valueOf(dagNode.getValue().getCurrentGeneration())); Properties configAsProperties = new Properties(jobSpec.getConfigAsProperties()); configAsProperties.putAll(configWithCurrentAttempts); //Return new spec with new config to avoid change the reference to dagNode return new JobSpec(jobSpec.getUri(), jobSpec.getVersion(), jobSpec.getDescription(), ConfigFactory.parseMap(configWithCurrentAttempts).withFallback(jobSpec.getConfig()), configAsProperties, jobSpec.getTemplateURI(), jobSpec.getJobTemplate(), jobSpec.getMetadata()); } static Config getJobConfig(DagNode<JobExecutionPlan> dagNode) { return dagNode.getValue().getJobSpec().getConfig(); } static SpecProducer<Spec> getSpecProducer(DagNode<JobExecutionPlan> dagNode) throws ExecutionException, InterruptedException { return dagNode.getValue().getSpecExecutor().getProducer().get(); } static ExecutionStatus getExecutionStatus(DagNode<JobExecutionPlan> dagNode) { return dagNode.getValue().getExecutionStatus(); } /** * Traverse the dag to determine the next set of nodes to be executed. It starts with the startNodes of the dag and * identifies each node yet to be executed and for which each of its parent nodes is in the {@link ExecutionStatus#COMPLETE} * state. */ static Set<DagNode<JobExecutionPlan>> getNext(Dag<JobExecutionPlan> dag) { Set<DagNode<JobExecutionPlan>> nextNodesToExecute = new HashSet<>(); LinkedList<DagNode<JobExecutionPlan>> nodesToExpand = Lists.newLinkedList(dag.getStartNodes()); FailureOption failureOption = getFailureOption(dag); while (!nodesToExpand.isEmpty()) { DagNode<JobExecutionPlan> node = nodesToExpand.poll(); ExecutionStatus executionStatus = getExecutionStatus(node); boolean addFlag = true; if (executionStatus == ExecutionStatus.PENDING || executionStatus == ExecutionStatus.PENDING_RETRY || executionStatus == ExecutionStatus.PENDING_RESUME) { //Add a node to be executed next, only if all of its parent nodes are COMPLETE. List<DagNode<JobExecutionPlan>> parentNodes = dag.getParents(node); for (DagNode<JobExecutionPlan> parentNode : parentNodes) { if (getExecutionStatus(parentNode) != ExecutionStatus.COMPLETE) { addFlag = false; break; } } if (addFlag) { nextNodesToExecute.add(node); } } else if (executionStatus == ExecutionStatus.COMPLETE) { //Explore the children of COMPLETED node as next candidates for execution. nodesToExpand.addAll(dag.getChildren(node)); } else if ((executionStatus == ExecutionStatus.FAILED) || (executionStatus == ExecutionStatus.CANCELLED)) { switch (failureOption) { case FINISH_RUNNING: return new HashSet<>(); case FINISH_ALL_POSSIBLE: default: break; } } } return nextNodesToExecute; } static FailureOption getFailureOption(Dag<JobExecutionPlan> dag) { if (dag.isEmpty()) { return null; } DagNode<JobExecutionPlan> dagNode = dag.getStartNodes().get(0); String failureOption = ConfigUtils.getString(getJobConfig(dagNode), ConfigurationKeys.FLOW_FAILURE_OPTION, DagManager.DEFAULT_FLOW_FAILURE_OPTION); return FailureOption.valueOf(failureOption); } static String getSpecExecutorUri(DagNode<JobExecutionPlan> dagNode) { return dagNode.getValue().getSpecExecutor().getUri().toString(); } static String getSerializedRequesterList(DagNode<JobExecutionPlan> dagNode) { return ConfigUtils.getString(dagNode.getValue().getJobSpec().getConfig(), RequesterService.REQUESTER_LIST, null); } static String getUserQuotaKey(String user, DagNode<JobExecutionPlan> dagNode) { return user + QUOTA_KEY_SEPERATOR + getSpecExecutorUri(dagNode); } static String getFlowGroupQuotaKey(String flowGroup, DagNode<JobExecutionPlan> dagNode) { return flowGroup + QUOTA_KEY_SEPERATOR + getSpecExecutorUri(dagNode); } /** * Increment the value of {@link JobExecutionPlan#currentAttempts} */ static void incrementJobAttempt(DagNode<JobExecutionPlan> dagNode) { dagNode.getValue().setCurrentAttempts(dagNode.getValue().getCurrentAttempts() + 1); } /** * Increment the value of {@link JobExecutionPlan#currentGeneration} * This method is not thread safe, we achieve correctness by making sure * one dag will only be handled in the same DagManagerThread */ static void incrementJobGeneration(DagNode<JobExecutionPlan> dagNode) { dagNode.getValue().setCurrentGeneration(dagNode.getValue().getCurrentGeneration() + 1); } /** * Flow start time is the same as the flow execution id which is the timestamp flow request was received, unless it * is a resumed flow, in which case it is {@link JobExecutionPlan#getFlowStartTime()} * @param dagNode dag node in context * @return flow start time */ static long getFlowStartTime(DagNode<JobExecutionPlan> dagNode) { long flowStartTime = dagNode.getValue().getFlowStartTime(); return flowStartTime == 0L ? getFlowExecId(dagNode) : flowStartTime; } /** * get the sla from the dag node config. * if time unit is not provided, it assumes time unit is minute. * @param dagNode dag node for which sla is to be retrieved * @return sla if it is provided, DEFAULT_FLOW_SLA_MILLIS otherwise */ static long getFlowSLA(DagNode<JobExecutionPlan> dagNode) { Config jobConfig = dagNode.getValue().getJobSpec().getConfig(); TimeUnit slaTimeUnit = TimeUnit.valueOf(ConfigUtils.getString( jobConfig, ConfigurationKeys.GOBBLIN_FLOW_SLA_TIME_UNIT, ConfigurationKeys.DEFAULT_GOBBLIN_FLOW_SLA_TIME_UNIT)); return jobConfig.hasPath(ConfigurationKeys.GOBBLIN_FLOW_SLA_TIME) ? slaTimeUnit.toMillis(jobConfig.getLong(ConfigurationKeys.GOBBLIN_FLOW_SLA_TIME)) : DEFAULT_FLOW_SLA_MILLIS; } /** * get the job start sla from the dag node config. * if time unit is not provided, it assumes time unit is minute. * @param dagNode dag node for which flow start sla is to be retrieved * @return job start sla in ms */ static long getJobStartSla(DagNode<JobExecutionPlan> dagNode, Long defaultJobStartSla) { Config jobConfig = dagNode.getValue().getJobSpec().getConfig(); TimeUnit slaTimeUnit = TimeUnit.valueOf(ConfigUtils.getString( jobConfig, ConfigurationKeys.GOBBLIN_JOB_START_SLA_TIME_UNIT, ConfigurationKeys.FALLBACK_GOBBLIN_JOB_START_SLA_TIME_UNIT)); return jobConfig.hasPath(ConfigurationKeys.GOBBLIN_JOB_START_SLA_TIME) ? slaTimeUnit.toMillis(jobConfig.getLong(ConfigurationKeys.GOBBLIN_JOB_START_SLA_TIME)) : defaultJobStartSla; } static int getDagQueueId(Dag<JobExecutionPlan> dag, int numThreads) { return getDagQueueId(DagManagerUtils.getFlowExecId(dag), numThreads); } static int getDagQueueId(long flowExecutionId, int numThreads) { return (int) (flowExecutionId % numThreads); } static Config getDagJobConfig(Dag<JobExecutionPlan> dag) { // Every dag should have at least one node, and the job configurations are cloned among each node return dag.getStartNodes().get(0).getValue().getJobSpec().getConfig(); } static boolean shouldFlowOutputMetrics(Dag<JobExecutionPlan> dag) { // defaults to false (so metrics are still tracked) if the dag property is not configured due to old dags return ConfigUtils.getBoolean(getDagJobConfig(dag), ConfigurationKeys.GOBBLIN_OUTPUT_JOB_LEVEL_METRICS, true); } static String getSpecExecutorName(DagNode<JobExecutionPlan> dagNode) { return dagNode.getValue().getSpecExecutor().getUri().toString(); } static void emitFlowEvent(Optional<EventSubmitter> eventSubmitter, Dag<JobExecutionPlan> dag, String flowEvent) { if (eventSubmitter.isPresent() && !dag.isEmpty()) { // Every dag node will contain the same flow metadata Config config = getDagJobConfig(dag); Map<String, String> flowMetadata = TimingEventUtils.getFlowMetadata(config); if (dag.getFlowEvent() != null) { flowEvent = dag.getFlowEvent(); } if (dag.getMessage() != null) { flowMetadata.put(TimingEvent.METADATA_MESSAGE, dag.getMessage()); } eventSubmitter.get().getTimingEvent(flowEvent).stop(flowMetadata); } } static List<String> getDistinctUniqueRequesters(String serializedRequesters) { if (serializedRequesters == null) { return Collections.emptyList(); } List<String> uniqueRequesters; try { uniqueRequesters = RequesterService.deserialize(serializedRequesters) .stream() .map(ServiceRequester::getName) .distinct() .collect(Collectors.toList()); return uniqueRequesters; } catch (IOException e) { throw new RuntimeException("Could not process requesters due to ", e); } } }
3,901
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/orchestration/InMemoryUserQuotaManager.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.orchestration; import com.google.inject.Inject; import com.typesafe.config.Config; import java.io.IOException; import java.util.Collection; import java.util.List; import java.util.Map; import java.util.Set; import java.util.concurrent.ConcurrentHashMap; import javax.inject.Singleton; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.exception.QuotaExceededException; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.util.ConfigUtils; import static org.apache.gobblin.service.ExecutionStatus.RUNNING; /** * An implementation of {@link UserQuotaManager} that stores quota usage in memory. */ @Slf4j @Singleton public class InMemoryUserQuotaManager extends AbstractUserQuotaManager { private final Map<String, Integer> proxyUserToJobCount = new ConcurrentHashMap<>(); private final Map<String, Integer> flowGroupToJobCount = new ConcurrentHashMap<>(); private final Map<String, Integer> requesterToJobCount = new ConcurrentHashMap<>(); private final Set<String> runningDagIds; @Inject public InMemoryUserQuotaManager(Config config) { super(config); this.runningDagIds = ConcurrentHashMap.newKeySet();; } protected QuotaCheck increaseAndCheckQuota(Dag.DagNode<JobExecutionPlan> dagNode) throws IOException { QuotaCheck quotaCheck = new QuotaCheck(true, true, true, ""); // Dag is already being tracked, no need to double increment for retries and multihop flows if (containsDagId(DagManagerUtils.generateDagId(dagNode).toString())) { return quotaCheck; } else { addDagId(DagManagerUtils.generateDagId(dagNode).toString()); } String proxyUser = ConfigUtils.getString(dagNode.getValue().getJobSpec().getConfig(), AzkabanProjectConfig.USER_TO_PROXY, null); String flowGroup = ConfigUtils.getString(dagNode.getValue().getJobSpec().getConfig(), ConfigurationKeys.FLOW_GROUP_KEY, ""); String specExecutorUri = DagManagerUtils.getSpecExecutorUri(dagNode); StringBuilder requesterMessage = new StringBuilder(); boolean proxyUserCheck; if (proxyUser != null && dagNode.getValue().getCurrentAttempts() <= 1) { int proxyQuotaIncrement = incrementJobCountAndCheckQuota( DagManagerUtils.getUserQuotaKey(proxyUser, dagNode), getQuotaForUser(proxyUser), CountType.USER_COUNT); proxyUserCheck = proxyQuotaIncrement >= 0; // proxy user quota check succeeds quotaCheck.setProxyUserCheck(proxyUserCheck); if (!proxyUserCheck) { // add 1 to proxyUserIncrement since proxyQuotaIncrement is the count before the increment requesterMessage.append(String.format( "Quota exceeded for proxy user %s on executor %s : quota=%s, requests above quota=%d%n", proxyUser, specExecutorUri, getQuotaForUser(proxyUser), Math.abs(proxyQuotaIncrement) + 1 - getQuotaForUser(proxyUser))); } } String serializedRequesters = DagManagerUtils.getSerializedRequesterList(dagNode); boolean requesterCheck = true; if (dagNode.getValue().getCurrentAttempts() <= 1) { List<String> uniqueRequesters = DagManagerUtils.getDistinctUniqueRequesters(serializedRequesters); for (String requester : uniqueRequesters) { int userQuotaIncrement = incrementJobCountAndCheckQuota( DagManagerUtils.getUserQuotaKey(requester, dagNode), getQuotaForUser(requester), CountType.REQUESTER_COUNT); boolean thisRequesterCheck = userQuotaIncrement >= 0; // user quota check succeeds requesterCheck = requesterCheck && thisRequesterCheck; quotaCheck.setRequesterCheck(requesterCheck); if (!thisRequesterCheck) { requesterMessage.append(String.format( "Quota exceeded for requester %s on executor %s : quota=%s, requests above quota=%d%n. ", requester, specExecutorUri, getQuotaForUser(requester), Math.abs(userQuotaIncrement) + 1 - getQuotaForUser(requester))); } } } boolean flowGroupCheck; if (dagNode.getValue().getCurrentAttempts() <= 1) { int flowGroupQuotaIncrement = incrementJobCountAndCheckQuota( DagManagerUtils.getFlowGroupQuotaKey(flowGroup, dagNode), getQuotaForFlowGroup(flowGroup), CountType.FLOWGROUP_COUNT); flowGroupCheck = flowGroupQuotaIncrement >= 0; quotaCheck.setFlowGroupCheck(flowGroupCheck); if (!flowGroupCheck) { requesterMessage.append(String.format("Quota exceeded for flowgroup %s on executor %s : quota=%s, requests above quota=%d%n", flowGroup, specExecutorUri, getQuotaForFlowGroup(flowGroup), Math.abs(flowGroupQuotaIncrement) + 1 - getQuotaForFlowGroup(flowGroup))); } } quotaCheck.setRequesterMessage(requesterMessage.toString()); return quotaCheck; } protected void rollbackIncrements(Dag.DagNode<JobExecutionPlan> dagNode) throws IOException { String proxyUser = ConfigUtils.getString(dagNode.getValue().getJobSpec().getConfig(), AzkabanProjectConfig.USER_TO_PROXY, null); String flowGroup = ConfigUtils.getString(dagNode.getValue().getJobSpec().getConfig(), ConfigurationKeys.FLOW_GROUP_KEY, ""); List<String> usersQuotaIncrement = DagManagerUtils.getDistinctUniqueRequesters(DagManagerUtils.getSerializedRequesterList(dagNode)); decrementJobCount(DagManagerUtils.getUserQuotaKey(proxyUser, dagNode), CountType.USER_COUNT); decrementQuotaUsageForUsers(usersQuotaIncrement); decrementJobCount(DagManagerUtils.getFlowGroupQuotaKey(flowGroup, dagNode), CountType.FLOWGROUP_COUNT); removeDagId(DagManagerUtils.generateDagId(dagNode).toString()); } private int incrementJobCountAndCheckQuota(String key, int keyQuota, CountType countType) throws IOException { int currentCount = incrementJobCount(key, countType); if (currentCount >= keyQuota) { return -currentCount; } else { return currentCount; } } private void decrementQuotaUsageForUsers(List<String> requestersToDecreaseCount) throws IOException { for (String requester : requestersToDecreaseCount) { decrementJobCount(requester, CountType.REQUESTER_COUNT); } } /** * Decrement the quota by one for the proxy user and requesters corresponding to the provided {@link Dag.DagNode}. * Returns true if the dag existed in the set of running dags and was removed successfully */ public boolean releaseQuota(Dag.DagNode<JobExecutionPlan> dagNode) throws IOException { boolean val = removeDagId(DagManagerUtils.generateDagId(dagNode).toString()); if (!val) { return false; } String proxyUser = ConfigUtils.getString(dagNode.getValue().getJobSpec().getConfig(), AzkabanProjectConfig.USER_TO_PROXY, null); if (proxyUser != null) { String proxyUserKey = DagManagerUtils.getUserQuotaKey(proxyUser, dagNode); decrementJobCount(proxyUserKey, CountType.USER_COUNT); } String flowGroup = ConfigUtils.getString(dagNode.getValue().getJobSpec().getConfig(), ConfigurationKeys.FLOW_GROUP_KEY, ""); decrementJobCount(DagManagerUtils.getFlowGroupQuotaKey(flowGroup, dagNode), CountType.FLOWGROUP_COUNT); String serializedRequesters = DagManagerUtils.getSerializedRequesterList(dagNode); try { for (String requester : DagManagerUtils.getDistinctUniqueRequesters(serializedRequesters)) { String requesterKey = DagManagerUtils.getUserQuotaKey(requester, dagNode); decrementJobCount(requesterKey, CountType.REQUESTER_COUNT); } } catch (IOException e) { log.error("Failed to release quota for requester list " + serializedRequesters, e); return false; } return true; } void addDagId(String dagId) { this.runningDagIds.add(dagId); } @Override boolean containsDagId(String dagId) { return this.runningDagIds.contains(dagId); } boolean removeDagId(String dagId) { return this.runningDagIds.remove(dagId); } public void init(Collection<Dag<JobExecutionPlan>> dags) throws IOException { for (Dag<JobExecutionPlan> dag : dags) { for (Dag.DagNode<JobExecutionPlan> dagNode : dag.getNodes()) { if (DagManagerUtils.getExecutionStatus(dagNode) == RUNNING) { // Add all the currently running Dags to the quota limit per user increaseAndCheckQuota(dagNode); } } } } public void checkQuota(Collection<Dag.DagNode<JobExecutionPlan>> dagNodes) throws IOException { for (Dag.DagNode<JobExecutionPlan> dagNode : dagNodes) { QuotaCheck quotaCheck = increaseAndCheckQuota(dagNode); if ((!quotaCheck.proxyUserCheck || !quotaCheck.requesterCheck || !quotaCheck.flowGroupCheck)) { // roll back the increased counts in this block rollbackIncrements(dagNode); throw new QuotaExceededException(quotaCheck.requesterMessage); } } } private int incrementJobCount(String key, Map<String, Integer> quotaMap) { Integer currentCount; // Modifications must be thread safe since DAGs on DagManagerThreads may update the quota for the same user do { currentCount = quotaMap.get(key); } while (currentCount == null ? quotaMap.putIfAbsent(key, 1) != null : !quotaMap.replace(key, currentCount, currentCount + 1)); if (currentCount == null) { currentCount = 0; } return currentCount; } private void decrementJobCount(String key, Map<String, Integer> quotaMap) { Integer currentCount; if (key == null) { return; } do { currentCount = quotaMap.get(key); } while (currentCount != null && currentCount > 0 && !quotaMap.replace(key, currentCount, currentCount - 1)); if (currentCount == null || currentCount == 0) { log.warn("Decrement job count was called for " + key + " when the count was already zero/absent."); } } int incrementJobCount(String user, CountType countType) throws IOException { switch (countType) { case USER_COUNT: return incrementJobCount(user, proxyUserToJobCount); case REQUESTER_COUNT: return incrementJobCount(user, requesterToJobCount); case FLOWGROUP_COUNT: return incrementJobCount(user, flowGroupToJobCount); default: throw new IOException("Invalid count type " + countType); } } void decrementJobCount(String user, CountType countType) throws IOException { switch (countType) { case USER_COUNT: decrementJobCount(user, proxyUserToJobCount); break; case REQUESTER_COUNT: decrementJobCount(user, requesterToJobCount); break; case FLOWGROUP_COUNT: decrementJobCount(user, flowGroupToJobCount); break; default: throw new IOException("Invalid count type " + countType); } } }
3,902
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/orchestration/Orchestrator.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.orchestration; import com.codahale.metrics.Counter; import com.codahale.metrics.Meter; import com.codahale.metrics.Timer; import com.google.common.annotations.VisibleForTesting; import com.google.common.base.Optional; import com.typesafe.config.Config; import java.io.IOException; import java.lang.reflect.InvocationTargetException; import java.net.URI; import java.util.Collections; import java.util.List; import java.util.Map; import java.util.Properties; import java.util.concurrent.TimeUnit; import javax.annotation.Nonnull; import javax.inject.Inject; import javax.inject.Singleton; import lombok.Getter; import lombok.Setter; import org.apache.commons.lang3.reflect.ConstructorUtils; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.configuration.State; import org.apache.gobblin.instrumented.Instrumentable; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.metrics.MetricContext; import org.apache.gobblin.metrics.ServiceMetricNames; import org.apache.gobblin.metrics.Tag; import org.apache.gobblin.metrics.event.EventSubmitter; import org.apache.gobblin.metrics.event.TimingEvent; import org.apache.gobblin.runtime.api.DagActionStore; import org.apache.gobblin.runtime.api.FlowSpec; import org.apache.gobblin.runtime.api.JobSpec; import org.apache.gobblin.runtime.api.Spec; import org.apache.gobblin.runtime.api.SpecCatalogListener; import org.apache.gobblin.runtime.api.SpecProducer; import org.apache.gobblin.runtime.api.TopologySpec; import org.apache.gobblin.runtime.spec_catalog.AddSpecResponse; import org.apache.gobblin.runtime.spec_catalog.TopologyCatalog; import org.apache.gobblin.service.ServiceConfigKeys; import org.apache.gobblin.service.modules.flow.SpecCompiler; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.service.modules.utils.FlowCompilationValidationHelper; import org.apache.gobblin.service.modules.utils.SharedFlowMetricsSingleton; import org.apache.gobblin.service.monitoring.FlowStatusGenerator; import org.apache.gobblin.util.ClassAliasResolver; import org.apache.gobblin.util.ConfigUtils; import org.apache.gobblin.util.reflection.GobblinConstructorUtils; import org.slf4j.Logger; import org.slf4j.LoggerFactory; /** * Orchestrator that is a {@link SpecCatalogListener}. It listens to changes * to {@link TopologyCatalog} and updates {@link SpecCompiler} state * Also it listens to {@link org.apache.gobblin.runtime.spec_catalog.FlowCatalog} and use the compiler to compile the new flow spec. */ @Alpha @Singleton public class Orchestrator implements SpecCatalogListener, Instrumentable { protected final Logger _log; protected final SpecCompiler specCompiler; protected final Optional<TopologyCatalog> topologyCatalog; protected final Optional<DagManager> dagManager; protected final MetricContext metricContext; protected final Optional<EventSubmitter> eventSubmitter; private final boolean isFlowConcurrencyEnabled; @Getter private Optional<Meter> flowOrchestrationSuccessFulMeter; @Getter private Optional<Meter> flowOrchestrationFailedMeter; @Getter private Optional<Timer> flowOrchestrationTimer; private Optional<Counter> flowFailedForwardToDagManagerCounter; @Setter private FlowStatusGenerator flowStatusGenerator; private UserQuotaManager quotaManager; private final FlowCompilationValidationHelper flowCompilationValidationHelper; private Optional<FlowTriggerHandler> flowTriggerHandler; @Getter private final SharedFlowMetricsSingleton sharedFlowMetricsSingleton; private final ClassAliasResolver<SpecCompiler> aliasResolver; public Orchestrator(Config config, Optional<TopologyCatalog> topologyCatalog, Optional<DagManager> dagManager, Optional<Logger> log, FlowStatusGenerator flowStatusGenerator, boolean instrumentationEnabled, Optional<FlowTriggerHandler> flowTriggerHandler, SharedFlowMetricsSingleton sharedFlowMetricsSingleton) { _log = log.isPresent() ? log.get() : LoggerFactory.getLogger(getClass()); this.aliasResolver = new ClassAliasResolver<>(SpecCompiler.class); this.topologyCatalog = topologyCatalog; this.dagManager = dagManager; this.flowStatusGenerator = flowStatusGenerator; this.flowTriggerHandler = flowTriggerHandler; this.sharedFlowMetricsSingleton = sharedFlowMetricsSingleton; try { String specCompilerClassName = ServiceConfigKeys.DEFAULT_GOBBLIN_SERVICE_FLOWCOMPILER_CLASS; if (config.hasPath(ServiceConfigKeys.GOBBLIN_SERVICE_FLOWCOMPILER_CLASS_KEY)) { specCompilerClassName = config.getString(ServiceConfigKeys.GOBBLIN_SERVICE_FLOWCOMPILER_CLASS_KEY); } _log.info("Using specCompiler class name/alias " + specCompilerClassName); this.specCompiler = (SpecCompiler) ConstructorUtils.invokeConstructor(Class.forName(this.aliasResolver.resolve(specCompilerClassName)), config); } catch (NoSuchMethodException | IllegalAccessException | InvocationTargetException | InstantiationException | ClassNotFoundException e) { throw new RuntimeException(e); } //At this point, the TopologySpecMap is initialized by the SpecCompiler. Pass the TopologySpecMap to the DagManager. if (this.dagManager.isPresent()) { this.dagManager.get().setTopologySpecMap(getSpecCompiler().getTopologySpecMap()); } if (instrumentationEnabled) { this.metricContext = Instrumented.getMetricContext(ConfigUtils.configToState(config), this.specCompiler.getClass()); this.flowOrchestrationSuccessFulMeter = Optional.of(this.metricContext.meter(ServiceMetricNames.FLOW_ORCHESTRATION_SUCCESSFUL_METER)); this.flowOrchestrationFailedMeter = Optional.of(this.metricContext.meter(ServiceMetricNames.FLOW_ORCHESTRATION_FAILED_METER)); this.flowOrchestrationTimer = Optional.of(this.metricContext.timer(ServiceMetricNames.FLOW_ORCHESTRATION_TIMER)); this.flowFailedForwardToDagManagerCounter = Optional.of(this.metricContext.counter(ServiceMetricNames.FLOW_FAILED_FORWARD_TO_DAG_MANAGER_COUNT)); this.eventSubmitter = Optional.of(new EventSubmitter.Builder(this.metricContext, "org.apache.gobblin.service").build()); } else { this.metricContext = null; this.flowOrchestrationSuccessFulMeter = Optional.absent(); this.flowOrchestrationFailedMeter = Optional.absent(); this.flowOrchestrationTimer = Optional.absent(); this.flowFailedForwardToDagManagerCounter = Optional.absent(); this.eventSubmitter = Optional.absent(); } this.isFlowConcurrencyEnabled = ConfigUtils.getBoolean(config, ServiceConfigKeys.FLOW_CONCURRENCY_ALLOWED, ServiceConfigKeys.DEFAULT_FLOW_CONCURRENCY_ALLOWED); quotaManager = GobblinConstructorUtils.invokeConstructor(UserQuotaManager.class, ConfigUtils.getString(config, ServiceConfigKeys.QUOTA_MANAGER_CLASS, ServiceConfigKeys.DEFAULT_QUOTA_MANAGER), config); this.flowCompilationValidationHelper = new FlowCompilationValidationHelper(sharedFlowMetricsSingleton, specCompiler, quotaManager, eventSubmitter, flowStatusGenerator, isFlowConcurrencyEnabled); } @Inject public Orchestrator(Config config, FlowStatusGenerator flowStatusGenerator, Optional<TopologyCatalog> topologyCatalog, Optional<DagManager> dagManager, Optional<Logger> log, Optional<FlowTriggerHandler> flowTriggerHandler, SharedFlowMetricsSingleton sharedFlowMetricsSingleton) { this(config, topologyCatalog, dagManager, log, flowStatusGenerator, true, flowTriggerHandler, sharedFlowMetricsSingleton); } @VisibleForTesting public SpecCompiler getSpecCompiler() { return this.specCompiler; } /** {@inheritDoc} */ @Override public AddSpecResponse onAddSpec(Spec addedSpec) { if (addedSpec instanceof TopologySpec) { _log.info("New Spec detected of type TopologySpec: " + addedSpec); this.specCompiler.onAddSpec(addedSpec); } else if (addedSpec instanceof FlowSpec) { _log.info("New Spec detected of type FlowSpec: " + addedSpec); return this.specCompiler.onAddSpec(addedSpec); } return new AddSpecResponse(null); } public void onDeleteSpec(URI deletedSpecURI, String deletedSpecVersion) { onDeleteSpec(deletedSpecURI, deletedSpecVersion, new Properties()); } /** {@inheritDoc} */ @Override public void onDeleteSpec(URI deletedSpecURI, String deletedSpecVersion, Properties headers) { _log.info("Spec deletion detected: " + deletedSpecURI + "/" + deletedSpecVersion); if (topologyCatalog.isPresent()) { this.specCompiler.onDeleteSpec(deletedSpecURI, deletedSpecVersion, headers); } } /** {@inheritDoc} */ @Override public void onUpdateSpec(Spec updatedSpec) { _log.info("Spec changed: " + updatedSpec); if (updatedSpec instanceof FlowSpec) { onAddSpec(updatedSpec); } if (!(updatedSpec instanceof TopologySpec)) { return; } try { onDeleteSpec(updatedSpec.getUri(), updatedSpec.getVersion()); } catch (Exception e) { _log.error("Failed to update Spec: " + updatedSpec, e); } try { onAddSpec(updatedSpec); } catch (Exception e) { _log.error("Failed to update Spec: " + updatedSpec, e); } } public void orchestrate(Spec spec, Properties jobProps, long triggerTimestampMillis, boolean isReminderEvent) throws Exception { // Add below waiting because TopologyCatalog and FlowCatalog service can be launched at the same time this.topologyCatalog.get().getInitComplete().await(); //Wait for the SpecCompiler to become healthy. this.getSpecCompiler().awaitHealthy(); long startTime = System.nanoTime(); if (spec instanceof FlowSpec) { Config flowConfig = ((FlowSpec) spec).getConfig(); String flowGroup = flowConfig.getString(ConfigurationKeys.FLOW_GROUP_KEY); String flowName = flowConfig.getString(ConfigurationKeys.FLOW_NAME_KEY); sharedFlowMetricsSingleton.addFlowGauge(spec, flowConfig, flowGroup, flowName); Optional<TimingEvent> flowCompilationTimer = this.eventSubmitter.transform(submitter -> new TimingEvent(submitter, TimingEvent.FlowTimings.FLOW_COMPILED)); Optional<Dag<JobExecutionPlan>> jobExecutionPlanDagOptional = this.flowCompilationValidationHelper.validateAndHandleConcurrentExecution(flowConfig, spec, flowGroup, flowName); if (!jobExecutionPlanDagOptional.isPresent()) { Instrumented.markMeter(this.flowOrchestrationFailedMeter); return; } Map<String, String> flowMetadata = TimingEventUtils.getFlowMetadata((FlowSpec) spec); FlowCompilationValidationHelper.addFlowExecutionIdIfAbsent(flowMetadata, jobExecutionPlanDagOptional.get()); // If multi-active scheduler is enabled do not pass onto DagManager, otherwise scheduler forwards it directly // Skip flow compilation as well, since we recompile after receiving event from DagActionStoreChangeMonitor later if (flowTriggerHandler.isPresent()) { // If triggerTimestampMillis was not set by the job trigger handler, then we do not handle this event if (triggerTimestampMillis == Long.parseLong(ConfigurationKeys.ORCHESTRATOR_TRIGGER_EVENT_TIME_NEVER_SET_VAL)) { _log.warn("Skipping execution of spec: {} because missing trigger timestamp in job properties", jobProps.getProperty(ConfigurationKeys.JOB_NAME_KEY)); flowMetadata.put(TimingEvent.METADATA_MESSAGE, "Flow orchestration skipped because no trigger timestamp " + "associated with flow action."); if (this.eventSubmitter.isPresent()) { new TimingEvent(this.eventSubmitter.get(), TimingEvent.FlowTimings.FLOW_FAILED).stop(flowMetadata); } return; } String flowExecutionId = flowMetadata.get(TimingEvent.FlowEventConstants.FLOW_EXECUTION_ID_FIELD); DagActionStore.DagAction flowAction = new DagActionStore.DagAction(flowGroup, flowName, flowExecutionId, DagActionStore.FlowActionType.LAUNCH); flowTriggerHandler.get().handleTriggerEvent(jobProps, flowAction, triggerTimestampMillis, isReminderEvent); _log.info("Multi-active scheduler finished handling trigger event: [{}, is: {}, triggerEventTimestamp: {}]", flowAction, isReminderEvent ? "reminder" : "original", triggerTimestampMillis); } else { Dag<JobExecutionPlan> jobExecutionPlanDag = jobExecutionPlanDagOptional.get(); if (jobExecutionPlanDag == null || jobExecutionPlanDag.isEmpty()) { FlowCompilationValidationHelper.populateFlowCompilationFailedEventMessage(eventSubmitter, spec, flowMetadata); Instrumented.markMeter(this.flowOrchestrationFailedMeter); sharedFlowMetricsSingleton.conditionallyUpdateFlowGaugeSpecState(spec, SharedFlowMetricsSingleton.CompiledState.FAILED); _log.warn("Cannot determine an executor to run on for Spec: " + spec); return; } sharedFlowMetricsSingleton.conditionallyUpdateFlowGaugeSpecState(spec, SharedFlowMetricsSingleton.CompiledState.SUCCESSFUL); FlowCompilationValidationHelper.addFlowExecutionIdIfAbsent(flowMetadata, jobExecutionPlanDag); if (flowCompilationTimer.isPresent()) { flowCompilationTimer.get().stop(flowMetadata); } // Depending on if DagManager is present, handle execution if (this.dagManager.isPresent()) { submitFlowToDagManager((FlowSpec) spec, jobExecutionPlanDag); } else { // Schedule all compiled JobSpecs on their respective Executor for (Dag.DagNode<JobExecutionPlan> dagNode : jobExecutionPlanDag.getNodes()) { DagManagerUtils.incrementJobAttempt(dagNode); JobExecutionPlan jobExecutionPlan = dagNode.getValue(); // Run this spec on selected executor SpecProducer producer = null; try { producer = jobExecutionPlan.getSpecExecutor().getProducer().get(); Spec jobSpec = jobExecutionPlan.getJobSpec(); if (!((JobSpec) jobSpec).getConfig().hasPath(ConfigurationKeys.FLOW_EXECUTION_ID_KEY)) { _log.warn("JobSpec does not contain flowExecutionId."); } Map<String, String> jobMetadata = TimingEventUtils.getJobMetadata(flowMetadata, jobExecutionPlan); _log.info(String.format("Going to orchestrate JobSpec: %s on Executor: %s", jobSpec, producer)); Optional<TimingEvent> jobOrchestrationTimer = this.eventSubmitter.transform( submitter -> new TimingEvent(submitter, TimingEvent.LauncherTimings.JOB_ORCHESTRATED)); producer.addSpec(jobSpec); if (jobOrchestrationTimer.isPresent()) { jobOrchestrationTimer.get().stop(jobMetadata); } } catch (Exception e) { _log.error("Cannot successfully setup spec: " + jobExecutionPlan.getJobSpec() + " on executor: " + producer + " for flow: " + spec, e); } } } } } else { Instrumented.markMeter(this.flowOrchestrationFailedMeter); throw new RuntimeException("Spec not of type FlowSpec, cannot orchestrate: " + spec); } Instrumented.markMeter(this.flowOrchestrationSuccessFulMeter); Instrumented.updateTimer(this.flowOrchestrationTimer, System.nanoTime() - startTime, TimeUnit.NANOSECONDS); } public void submitFlowToDagManager(FlowSpec flowSpec, Optional<String> optionalFlowExecutionId) throws IOException, InterruptedException { Optional<Dag<JobExecutionPlan>> optionalJobExecutionPlanDag = this.flowCompilationValidationHelper.createExecutionPlanIfValid(flowSpec, optionalFlowExecutionId); if (optionalJobExecutionPlanDag.isPresent()) { submitFlowToDagManager(flowSpec, optionalJobExecutionPlanDag.get()); } else { _log.warn("Flow: {} submitted to dagManager failed to compile and produce a job execution plan dag", flowSpec); Instrumented.markMeter(this.flowOrchestrationFailedMeter); } } public void submitFlowToDagManager(FlowSpec flowSpec, Dag<JobExecutionPlan> jobExecutionPlanDag) throws IOException { try { //Send the dag to the DagManager. this.dagManager.get().addDag(jobExecutionPlanDag, true, true); } catch (Exception ex) { String failureMessage = "Failed to add Job Execution Plan due to: " + ex.getMessage(); _log.warn("Orchestrator call - " + failureMessage, ex); if (this.flowFailedForwardToDagManagerCounter.isPresent()) { this.flowFailedForwardToDagManagerCounter.get().inc(); } if (this.eventSubmitter.isPresent()) { // pronounce failed before stack unwinds, to ensure flow not marooned in `COMPILED` state; (failure likely attributable to DB connection/failover) Map<String, String> flowMetadata = TimingEventUtils.getFlowMetadata(flowSpec); flowMetadata.put(TimingEvent.METADATA_MESSAGE, failureMessage); new TimingEvent(this.eventSubmitter.get(), TimingEvent.FlowTimings.FLOW_FAILED).stop(flowMetadata); } throw ex; } } public void remove(Spec spec, Properties headers) throws IOException { // TODO: Evolve logic to cache and reuse previously compiled JobSpecs // .. this will work for Identity compiler but not always for multi-hop. // Note: Current logic assumes compilation is consistent between all executions if (spec instanceof FlowSpec) { //Send the dag to the DagManager to stop it. //Also send it to the SpecProducer to do any cleanup tasks on SpecExecutor. if (this.dagManager.isPresent()) { _log.info("Forwarding cancel request for flow URI {} to DagManager.", spec.getUri()); this.dagManager.get().stopDag(spec.getUri()); } // We need to recompile the flow to find the spec producer, // If compilation result is different, its remove request can go to some different spec producer deleteFromExecutor(spec, headers); } else { throw new RuntimeException("Spec not of type FlowSpec, cannot delete: " + spec); } } private void deleteFromExecutor(Spec spec, Properties headers) { Dag<JobExecutionPlan> jobExecutionPlanDag = specCompiler.compileFlow(spec); if (jobExecutionPlanDag.isEmpty()) { _log.warn("Cannot determine an executor to delete Spec: " + spec); return; } // Delete all compiled JobSpecs on their respective Executor for (Dag.DagNode<JobExecutionPlan> dagNode : jobExecutionPlanDag.getNodes()) { JobExecutionPlan jobExecutionPlan = dagNode.getValue(); JobSpec jobSpec = jobExecutionPlan.getJobSpec(); try { SpecProducer<Spec> producer = jobExecutionPlan.getSpecExecutor().getProducer().get(); if (jobSpec.getConfig().hasPath(ConfigurationKeys.FLOW_EXECUTION_ID_KEY)) { headers.setProperty(ConfigurationKeys.FLOW_EXECUTION_ID_KEY, jobSpec.getConfig().getString(ConfigurationKeys.FLOW_EXECUTION_ID_KEY)); } _log.info(String.format("Going to delete JobSpec: %s on Executor: %s", jobSpec, producer)); producer.deleteSpec(jobSpec.getUri(), headers); } catch (Exception e) { _log.error(String.format("Could not delete JobSpec: %s for flow: %s", jobSpec, spec), e); } } } @Nonnull @Override public MetricContext getMetricContext() { return this.metricContext; } @Override public boolean isInstrumentationEnabled() { return null != this.metricContext; } @Override public List<Tag<?>> generateTags(State state) { return Collections.emptyList(); } @Override public void switchMetricContext(List<Tag<?>> tags) { throw new UnsupportedOperationException(); } @Override public void switchMetricContext(MetricContext context) { throw new UnsupportedOperationException(); } }
3,903
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/orchestration/MysqlDagStateStore.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.orchestration; import java.io.IOException; import java.lang.reflect.Type; import java.net.URI; import java.util.List; import java.util.Map; import java.util.Set; import java.util.stream.Collectors; import org.apache.gobblin.configuration.State; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.metastore.MysqlDagStateStoreFactory; import org.apache.gobblin.metastore.MysqlStateStore; import org.apache.gobblin.metastore.MysqlStateStoreEntryManager; import org.apache.gobblin.metastore.StateStore; import org.apache.gobblin.metastore.predicates.StateStorePredicate; import org.apache.gobblin.metrics.ContextAwareCounter; import org.apache.gobblin.metrics.MetricContext; import org.apache.gobblin.metrics.ServiceMetricNames; import org.apache.gobblin.runtime.api.TopologySpec; import org.apache.gobblin.runtime.spec_serde.GsonSerDe; import org.apache.gobblin.service.ServiceConfigKeys; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.service.modules.spec.JobExecutionPlanDagFactory; import org.apache.gobblin.service.modules.spec.JobExecutionPlanListDeserializer; import org.apache.gobblin.service.modules.spec.JobExecutionPlanListSerializer; import org.apache.gobblin.util.ConfigUtils; import com.google.common.base.Joiner; import com.google.common.base.Predicates; import com.google.gson.JsonDeserializer; import com.google.gson.JsonSerializer; import com.google.gson.reflect.TypeToken; import com.typesafe.config.Config; import static org.apache.gobblin.service.ServiceConfigKeys.GOBBLIN_SERVICE_PREFIX; import static org.apache.gobblin.service.modules.orchestration.DagManagerUtils.generateDagId; /** * A implementation of {@link DagStateStore} using MySQL as a backup, leverage {@link MysqlStateStore}. * It implements interfaces of {@link DagStateStore} but delegating responsibilities to methods provided * in {@link MysqlStateStore}. * It also implements conversion between {@link Dag<JobExecutionPlan>} to {@link State}. * * The schema of this will simply be: * | storeName | tableName | State | * where storeName represents FlowId, a combination of FlowGroup and FlowName, and tableName represents FlowExecutionId. * State is a pocket for serialized {@link Dag} object. * * */ public class MysqlDagStateStore implements DagStateStore { public static final String CONFIG_PREFIX = GOBBLIN_SERVICE_PREFIX + "mysqlDagStateStore"; public static final String DAG_KEY_IN_STATE = "dag"; /** * The schema of {@link MysqlStateStore} is fixed but the columns are semantically projected into Dag's context: * - The 'storeName' is FlowId. * - The 'tableName' is FlowExecutionId. */ private MysqlStateStore<State> mysqlStateStore; private final GsonSerDe<List<JobExecutionPlan>> serDe; private JobExecutionPlanDagFactory jobExecPlanDagFactory; private MetricContext metricContext; private ContextAwareCounter totalDagCount; public MysqlDagStateStore(Config config, Map<URI, TopologySpec> topologySpecMap) { if (config.hasPath(CONFIG_PREFIX)) { config = config.getConfig(CONFIG_PREFIX).withFallback(config); } this.mysqlStateStore = (MysqlStateStore<State>) createStateStore(config); JsonSerializer<List<JobExecutionPlan>> serializer = new JobExecutionPlanListSerializer(); JsonDeserializer<List<JobExecutionPlan>> deserializer = new JobExecutionPlanListDeserializer(topologySpecMap); Type typeToken = new TypeToken<List<JobExecutionPlan>>() { }.getType(); this.serDe = new GsonSerDe<>(typeToken, serializer, deserializer); this.jobExecPlanDagFactory = new JobExecutionPlanDagFactory(); this.metricContext = Instrumented.getMetricContext(new org.apache.gobblin.configuration.State(ConfigUtils.configToProperties(config)), this.getClass()); this.totalDagCount = this.metricContext.contextAwareCounter(ServiceMetricNames.DAG_COUNT_MYSQL_DAG_STATE_COUNT); } /** * Creating an instance of StateStore. */ protected StateStore<State> createStateStore(Config config) { try { return (MysqlDagStateStoreFactory.class.newInstance()).createStateStore(config, State.class); } catch (ReflectiveOperationException rfoe) { throw new RuntimeException("A MySQL StateStore cannot be correctly initialized due to:", rfoe); } } @Override public void writeCheckpoint(Dag<JobExecutionPlan> dag) throws IOException { mysqlStateStore.put(getStoreNameFromDagId(generateDagId(dag).toString()), getTableNameFromDagId(generateDagId(dag).toString()), convertDagIntoState(dag)); this.totalDagCount.inc(); } @Override public void cleanUp(Dag<JobExecutionPlan> dag) throws IOException { cleanUp(generateDagId(dag).toString()); } @Override public void cleanUp(String dagId) throws IOException { mysqlStateStore.delete(getStoreNameFromDagId(dagId), getTableNameFromDagId(dagId)); this.totalDagCount.dec(); } @Override public List<Dag<JobExecutionPlan>> getDags() throws IOException { return mysqlStateStore.getAll().stream().map(this::convertStateObjIntoDag).collect(Collectors.toList()); } @Override public Dag<JobExecutionPlan> getDag(String dagId) throws IOException { List<State> states = mysqlStateStore.getAll(getStoreNameFromDagId(dagId), getTableNameFromDagId(dagId)); if (states.isEmpty()) { return null; } return convertStateObjIntoDag(states.get(0)); } @Override public Set<String> getDagIds() throws IOException { List<MysqlStateStoreEntryManager> entries = (List<MysqlStateStoreEntryManager>) mysqlStateStore .getMetadataForTables(new StateStorePredicate(Predicates.alwaysTrue())); return entries.stream().map(entry -> entryToDagId(entry.getStoreName(), entry.getTableName())).collect(Collectors.toSet()); } /** * Convert a state store entry into a dag ID * e.g. storeName = group1_name1, tableName = 1234 gives dagId group1_name1_1234 */ private String entryToDagId(String storeName, String tableName) { return Joiner.on(ServiceConfigKeys.DAG_STORE_KEY_SEPARATION_CHARACTER).join(storeName, tableName); } /** * Return a storeName given a dagId. Store name is defined as flowGroup_flowName. */ private String getStoreNameFromDagId(String dagId) { return dagId.substring(0, dagId.lastIndexOf(ServiceConfigKeys.DAG_STORE_KEY_SEPARATION_CHARACTER)); } /** * Return a tableName given a dagId. Table name is defined as the flowExecutionId. */ private String getTableNameFromDagId(String dagId) { return dagId.substring(dagId.lastIndexOf(ServiceConfigKeys.DAG_STORE_KEY_SEPARATION_CHARACTER) + 1); } /** * For {@link Dag} to work with {@link MysqlStateStore}, it needs to be packaged into a {@link State} object. * The way that it does is simply serialize the {@link Dag} first and use the key {@link #DAG_KEY_IN_STATE} * to be pair with it. * * The serialization step is required for readability and portability of serde lib. * @param dag The dag to be converted. * @return An {@link State} object that contains a single k-v pair for {@link Dag}. */ private State convertDagIntoState(Dag<JobExecutionPlan> dag) { State outputState = new State(); // Make sure the object has been serialized. List<JobExecutionPlan> jobExecutionPlanList = dag.getNodes().stream().map(Dag.DagNode::getValue).collect(Collectors.toList()); outputState.setProp(DAG_KEY_IN_STATE, serDe.serialize(jobExecutionPlanList)); return outputState; } /** * Get the {@link Dag} out of a {@link State} pocket. */ private Dag<JobExecutionPlan> convertStateObjIntoDag(State state) { String serializedJobExecPlanList = state.getProp(DAG_KEY_IN_STATE); return jobExecPlanDagFactory.createDag(serDe.deserialize(serializedJobExecPlanList)); } }
3,904
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/orchestration/TimingEventUtils.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.orchestration; import java.util.Map; import com.google.common.collect.Maps; import com.typesafe.config.Config; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.metrics.event.TimingEvent; import org.apache.gobblin.runtime.api.FlowSpec; import org.apache.gobblin.runtime.api.JobSpec; import org.apache.gobblin.runtime.api.SpecExecutor; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.util.ConfigUtils; public class TimingEventUtils { public static Map<String, String> getFlowMetadata(FlowSpec flowSpec) { return getFlowMetadata(flowSpec.getConfig()); } static Map<String, String> getFlowMetadata(Config flowConfig) { Map<String, String> metadata = Maps.newHashMap(); metadata.put(TimingEvent.FlowEventConstants.FLOW_NAME_FIELD, flowConfig.getString(ConfigurationKeys.FLOW_NAME_KEY)); metadata.put(TimingEvent.FlowEventConstants.FLOW_GROUP_FIELD, flowConfig.getString(ConfigurationKeys.FLOW_GROUP_KEY)); if (flowConfig.hasPath(ConfigurationKeys.FLOW_EXECUTION_ID_KEY)) { metadata.put(TimingEvent.FlowEventConstants.FLOW_EXECUTION_ID_FIELD, flowConfig.getString(ConfigurationKeys.FLOW_EXECUTION_ID_KEY)); } return metadata; } static Map<String, String> getJobMetadata(Map<String, String> flowMetadata, JobExecutionPlan jobExecutionPlan) { Map<String, String> jobMetadata = Maps.newHashMap(); JobSpec jobSpec = jobExecutionPlan.getJobSpec(); SpecExecutor specExecutor = jobExecutionPlan.getSpecExecutor(); jobMetadata.putAll(flowMetadata); jobMetadata.put(TimingEvent.FlowEventConstants.FLOW_NAME_FIELD, jobSpec.getConfig().getString(ConfigurationKeys.FLOW_NAME_KEY)); jobMetadata.put(TimingEvent.FlowEventConstants.FLOW_GROUP_FIELD, jobSpec.getConfig().getString(ConfigurationKeys.FLOW_GROUP_KEY)); jobMetadata.put(TimingEvent.FlowEventConstants.FLOW_EXECUTION_ID_FIELD, jobSpec.getConfig().getString(ConfigurationKeys.FLOW_EXECUTION_ID_KEY)); jobMetadata.put(TimingEvent.FlowEventConstants.JOB_NAME_FIELD, jobSpec.getConfig().getString(ConfigurationKeys.JOB_NAME_KEY)); jobMetadata.put(TimingEvent.FlowEventConstants.JOB_GROUP_FIELD, jobSpec.getConfig().getString(ConfigurationKeys.JOB_GROUP_KEY)); jobMetadata.put(TimingEvent.FlowEventConstants.JOB_TAG_FIELD, ConfigUtils.getString(jobSpec.getConfig(), ConfigurationKeys.JOB_TAG_KEY, null)); jobMetadata.put(TimingEvent.FlowEventConstants.SPEC_EXECUTOR_FIELD, specExecutor.getUri().toString()); jobMetadata.put(TimingEvent.FlowEventConstants.MAX_ATTEMPTS_FIELD, Integer.toString(jobExecutionPlan.getMaxAttempts())); jobMetadata.put(TimingEvent.FlowEventConstants.CURRENT_ATTEMPTS_FIELD, Integer.toString(jobExecutionPlan.getCurrentAttempts())); jobMetadata.put(TimingEvent.FlowEventConstants.CURRENT_GENERATION_FIELD, Integer.toString(jobExecutionPlan.getCurrentGeneration())); jobMetadata.put(TimingEvent.FlowEventConstants.SHOULD_RETRY_FIELD, Boolean.toString(false)); jobMetadata.put(TimingEvent.FlowEventConstants.FLOW_EDGE_FIELD, ConfigUtils.getString(jobSpec.getConfig(), ConfigurationKeys.FLOW_EDGE_ID_KEY, "")); jobMetadata.put(TimingEvent.FlowEventConstants.FLOW_MODIFICATION_TIME_FIELD, Long.toString( ConfigUtils.getLong(jobSpec.getConfig(), FlowSpec.MODIFICATION_TIME_KEY, 0L))); return jobMetadata; } }
3,905
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/orchestration/FSDagStateStore.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.orchestration; import java.io.File; import java.io.IOException; import java.lang.reflect.Type; import java.net.URI; import java.util.HashSet; import java.util.List; import java.util.Map; import java.util.Set; import java.util.stream.Collectors; import org.apache.commons.lang3.StringUtils; import com.google.common.annotations.VisibleForTesting; import com.google.common.base.Charsets; import com.google.common.collect.Lists; import com.google.common.io.Files; import com.google.gson.JsonDeserializer; import com.google.gson.JsonSerializer; import com.google.gson.reflect.TypeToken; import com.typesafe.config.Config; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.metrics.ContextAwareCounter; import org.apache.gobblin.metrics.MetricContext; import org.apache.gobblin.metrics.ServiceMetricNames; import org.apache.gobblin.runtime.api.TopologySpec; import org.apache.gobblin.runtime.spec_serde.GsonSerDe; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.service.modules.spec.JobExecutionPlanDagFactory; import org.apache.gobblin.service.modules.spec.JobExecutionPlanListDeserializer; import org.apache.gobblin.service.modules.spec.JobExecutionPlanListSerializer; import org.apache.gobblin.util.ConfigUtils; @Alpha @Slf4j public class FSDagStateStore implements DagStateStore { public static final String DAG_FILE_EXTENSION = ".dag"; static final String DAG_STATESTORE_DIR = DagManager.DAG_MANAGER_PREFIX + "dagStateStoreDir"; private final String dagCheckpointDir; private final GsonSerDe<List<JobExecutionPlan>> serDe; private MetricContext metricContext; private ContextAwareCounter totalDagCount; public FSDagStateStore(Config config, Map<URI, TopologySpec> topologySpecMap) throws IOException { this.dagCheckpointDir = config.getString(DAG_STATESTORE_DIR); File checkpointDir = new File(this.dagCheckpointDir); if (!checkpointDir.exists()) { if (!checkpointDir.mkdirs()) { throw new IOException("Could not create dag state store dir - " + this.dagCheckpointDir); } } JsonSerializer<List<JobExecutionPlan>> serializer = new JobExecutionPlanListSerializer(); JsonDeserializer<List<JobExecutionPlan>> deserializer = new JobExecutionPlanListDeserializer(topologySpecMap); /** {@link Type} object will need to strictly match with the generic arguments being used * to define {@link GsonSerDe} * Due to type erasure, the {@link Type} needs to initialized here instead of inside {@link GsonSerDe}. * */ Type typeToken = new TypeToken<List<JobExecutionPlan>>(){}.getType(); this.serDe = new GsonSerDe<>(typeToken, serializer, deserializer); this.metricContext = Instrumented.getMetricContext(new org.apache.gobblin.configuration.State(ConfigUtils.configToProperties(config)), this.getClass()); this.totalDagCount = this.metricContext.contextAwareCounter(ServiceMetricNames.DAG_COUNT_FS_DAG_STATE_COUNT); } /** * {@inheritDoc} */ @Override public synchronized void writeCheckpoint(Dag<JobExecutionPlan> dag) throws IOException { // write to a temporary name then rename to make the operation atomic when the file system allows a file to be // replaced String fileName = DagManagerUtils.generateDagId(dag) + DAG_FILE_EXTENSION; String serializedDag = serializeDag(dag); File tmpCheckpointFile = new File(this.dagCheckpointDir, fileName + ".tmp"); File checkpointFile = new File(this.dagCheckpointDir, fileName); Files.write(serializedDag, tmpCheckpointFile, Charsets.UTF_8); Files.move(tmpCheckpointFile, checkpointFile); this.totalDagCount.inc(); } /** * {@inheritDoc} */ @Override public synchronized void cleanUp(Dag<JobExecutionPlan> dag) { cleanUp(DagManagerUtils.generateDagId(dag).toString()); } /** * {@inheritDoc} */ @Override public synchronized void cleanUp(String dagId) { String fileName = dagId + DAG_FILE_EXTENSION; //Delete the dag checkpoint file from the checkpoint directory File checkpointFile = new File(this.dagCheckpointDir, fileName); if (!checkpointFile.delete()) { log.error("Could not delete checkpoint file: {}", checkpointFile.getName()); } else { this.totalDagCount.dec(); } } /** * {@inheritDoc} */ @Override public List<Dag<JobExecutionPlan>> getDags() throws IOException { List<Dag<JobExecutionPlan>> runningDags = Lists.newArrayList(); File dagCheckpointFolder = new File(this.dagCheckpointDir); for (File file : dagCheckpointFolder.listFiles((dir, name) -> name.endsWith(DAG_FILE_EXTENSION))) { runningDags.add(getDag(file)); } return runningDags; } /** * {@inheritDoc} */ @Override public Dag<JobExecutionPlan> getDag(String dagId) throws IOException { File file = new File(this.dagCheckpointDir, dagId + DAG_FILE_EXTENSION); if (!file.exists()) { return null; } return getDag(file); } /** * Return a {@link Dag} given a file name. * @param dagFile * @return the {@link Dag} associated with the dagFile. */ @VisibleForTesting public Dag<JobExecutionPlan> getDag(File dagFile) throws IOException { String serializedDag = Files.toString(dagFile, Charsets.UTF_8); return deserializeDag(serializedDag); } /** * {@inheritDoc} */ @Override public Set<String> getDagIds() { Set<String> dagIds = new HashSet<>(); File dagCheckpointFolder = new File(this.dagCheckpointDir); for (File file : dagCheckpointFolder.listFiles((dir, name) -> name.endsWith(DAG_FILE_EXTENSION))) { dagIds.add(StringUtils.removeEnd(file.getName(), DAG_FILE_EXTENSION)); } return dagIds; } /** * Serialize a {@link Dag<JobExecutionPlan>}. * @param dag A Dag parametrized by type {@link JobExecutionPlan}. * @return a JSON string representation of the Dag object. */ private String serializeDag(Dag<JobExecutionPlan> dag) { List<JobExecutionPlan> jobExecutionPlanList = dag.getNodes().stream().map(Dag.DagNode::getValue).collect(Collectors.toList()); return serDe.serialize(jobExecutionPlanList); } /** * De-serialize a Dag. * @param jsonDag A string representation of a Dag. * @return a {@link Dag} parametrized by {@link JobExecutionPlan}. */ private Dag<JobExecutionPlan> deserializeDag(String jsonDag) { return new JobExecutionPlanDagFactory().createDag(serDe.deserialize(jsonDag)); } }
3,906
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/orchestration/AbstractUserQuotaManager.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.orchestration; import java.io.IOException; import java.util.Map; import com.google.common.collect.ImmutableMap; import com.typesafe.config.Config; import lombok.AllArgsConstructor; import lombok.Setter; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.metrics.MetricContext; import org.apache.gobblin.util.ConfigUtils; /** * An abstract implementation of {@link UserQuotaManager} that has base implementation of checking quota and increasing/decreasing it. */ @Slf4j abstract public class AbstractUserQuotaManager implements UserQuotaManager { public static final String PER_USER_QUOTA = DagManager.DAG_MANAGER_PREFIX + "perUserQuota"; public static final String PER_FLOWGROUP_QUOTA = DagManager.DAG_MANAGER_PREFIX + "perFlowGroupQuota"; public static final String USER_JOB_QUOTA_KEY = DagManager.DAG_MANAGER_PREFIX + "defaultJobQuota"; public static final String QUOTA_SEPERATOR = ":"; public static final Integer DEFAULT_USER_JOB_QUOTA = Integer.MAX_VALUE; private final Map<String, Integer> perUserQuota; private final Map<String, Integer> perFlowGroupQuota; protected MetricContext metricContext; private final int defaultQuota; public AbstractUserQuotaManager(Config config) { this.metricContext = Instrumented.getMetricContext(new org.apache.gobblin.configuration.State(ConfigUtils.configToProperties(config)), this.getClass()); this.defaultQuota = ConfigUtils.getInt(config, USER_JOB_QUOTA_KEY, DEFAULT_USER_JOB_QUOTA); ImmutableMap.Builder<String, Integer> userMapBuilder = ImmutableMap.builder(); ImmutableMap.Builder<String, Integer> flowGroupMapBuilder = ImmutableMap.builder(); // Quotas will take form of user:<Quota> and flowGroup:<Quota> for (String flowGroupQuota : ConfigUtils.getStringList(config, PER_FLOWGROUP_QUOTA)) { flowGroupMapBuilder.put(flowGroupQuota.split(QUOTA_SEPERATOR)[0], Integer.parseInt(flowGroupQuota.split(QUOTA_SEPERATOR)[1])); } // Keep quotas per user as well in form user:<Quota> which apply for all flowgroups for (String userQuota : ConfigUtils.getStringList(config, PER_USER_QUOTA)) { userMapBuilder.put(userQuota.split(QUOTA_SEPERATOR)[0], Integer.parseInt(userQuota.split(QUOTA_SEPERATOR)[1])); } this.perUserQuota = userMapBuilder.build(); this.perFlowGroupQuota = flowGroupMapBuilder.build(); } abstract boolean containsDagId(String dagId) throws IOException; int getQuotaForUser(String user) { return this.perUserQuota.getOrDefault(user, defaultQuota); } int getQuotaForFlowGroup(String flowGroup) { return this.perFlowGroupQuota.getOrDefault(flowGroup, defaultQuota); } @Setter @AllArgsConstructor protected static class QuotaCheck { boolean proxyUserCheck; boolean requesterCheck; boolean flowGroupCheck; String requesterMessage; } protected enum CountType { USER_COUNT, REQUESTER_COUNT, FLOWGROUP_COUNT } }
3,907
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/orchestration/UserQuotaManager.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.orchestration; import java.io.IOException; import java.util.Collection; import org.apache.gobblin.exception.QuotaExceededException; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; /** * Manages the statically configured user quotas for the proxy user in user.to.proxy configuration, the API requester(s) * and the flow group. * It is used by the {@link DagManager} to ensure that the number of currently running jobs do not exceed the quota, if * the quota is exceeded, the execution will fail without running on the underlying executor. */ public interface UserQuotaManager { /** * Initialize with the provided set of dags. */ void init(Collection<Dag<JobExecutionPlan>> dags) throws IOException; /** * Checks if the dagNode exceeds the statically configured user quota for the proxy user, requester user and flowGroup. * It also increases the quota usage for proxy user, requester and the flowGroup of the given DagNode by one. * @throws QuotaExceededException if the quota is exceeded */ void checkQuota(Collection<Dag.DagNode<JobExecutionPlan>> dagNode) throws IOException; /** * Decrement the quota by one for the proxy user and requesters corresponding to the provided {@link Dag.DagNode}. * Returns true if successfully reduces the quota usage */ boolean releaseQuota(Dag.DagNode<JobExecutionPlan> dagNode) throws IOException; }
3,908
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/db/ServiceDatabaseManager.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.db; import java.util.Objects; import org.flywaydb.core.Flyway; import com.google.common.util.concurrent.AbstractIdleService; import javax.inject.Inject; import javax.inject.Singleton; import lombok.extern.slf4j.Slf4j; /** * This class initializes and migrates the database schema for Gobblin service. * * We use Flyway to run the migrations that are defined in resources/org/apache/gobblin/service/db/migration * */ @Singleton @Slf4j public class ServiceDatabaseManager extends AbstractIdleService { private final ServiceDatabaseProvider databaseProvider; @Inject public ServiceDatabaseManager(ServiceDatabaseProvider databaseProvider) { this.databaseProvider = Objects.requireNonNull(databaseProvider); } @Override protected void startUp() throws Exception { Flyway flyway = Flyway.configure().locations("classpath:org/apache/gobblin/service/db/migration").failOnMissingLocations(true) .dataSource(databaseProvider.getDatasource()) // Existing GaaS DBs have state store tables. // Flyway will refuse to use such non-empty DBs by default. With baselineOnMigrate(true), it should // create new tables, while keeping old ones intact. .baselineOnMigrate(true) .baselineVersion("0") .load(); log.info("Ensuring service database is migrated to latest schema"); // Start the migration flyway.migrate(); } @Override protected void shutDown() throws Exception { } }
3,909
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/db/ServiceDatabaseProvider.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.db; import javax.sql.DataSource; /** * Provides access to Gobblin service database. * * DB schema is defined using Flyway migrations. * */ public interface ServiceDatabaseProvider { DataSource getDatasource(); }
3,910
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/db/ServiceDatabaseProviderImpl.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.db; import java.time.Duration; import java.util.Objects; import java.util.concurrent.atomic.AtomicInteger; import org.slf4j.Logger; import org.slf4j.LoggerFactory; import com.typesafe.config.Config; import com.zaxxer.hikari.HikariDataSource; import javax.inject.Inject; import javax.sql.DataSource; import lombok.AllArgsConstructor; import lombok.Builder; import lombok.Getter; import lombok.NoArgsConstructor; import org.apache.gobblin.util.jdbc.MysqlDataSourceUtils; import org.apache.gobblin.password.PasswordManager; import org.apache.gobblin.service.ServiceConfigKeys; import org.apache.gobblin.util.ConfigUtils; public class ServiceDatabaseProviderImpl implements ServiceDatabaseProvider { private static final Logger LOG = LoggerFactory.getLogger(ServiceDatabaseProviderImpl.class); private static final AtomicInteger POOL_NUM = new AtomicInteger(0); private final Configuration configuration; private HikariDataSource dataSource; @Inject public ServiceDatabaseProviderImpl(Configuration configuration) { this.configuration = configuration; } public DataSource getDatasource() { ensureDataSource(); return dataSource; } private synchronized void ensureDataSource() { if (dataSource != null) { return; } dataSource = new HikariDataSource(); dataSource.setPoolName("HikariPool-" + POOL_NUM.incrementAndGet() + "-" + getClass().getSimpleName()); dataSource.setJdbcUrl(configuration.getUrl()); dataSource.setUsername(configuration.getUserName()); dataSource.setPassword(configuration.getPassword()); // MySQL server can timeout a connection so need to validate connections before use final String validationQuery = MysqlDataSourceUtils.QUERY_CONNECTION_IS_VALID_AND_NOT_READONLY; LOG.info("setting `DataSource` validation query: '" + validationQuery + "'"); // TODO: revisit following verification of successful connection pool migration: // If your driver supports JDBC4 we strongly recommend not setting this property. This is for "legacy" drivers // that do not support the JDBC4 Connection.isValid() API; see: // https://github.com/brettwooldridge/HikariCP#gear-configuration-knobs-baby dataSource.setConnectionTestQuery(validationQuery); dataSource.setValidationTimeout(Duration.ofSeconds(5).toMillis()); // To improve performance, we set a maximum connection lifetime // If database goes to read-only mode, then connection would not work correctly for up to configured lifetime dataSource.setMaxLifetime(configuration.getMaxConnectionLifetime().toMillis()); dataSource.setIdleTimeout(Duration.ofSeconds(10).toMillis()); dataSource.setMinimumIdle(2); dataSource.setMaximumPoolSize(configuration.getMaxConnections()); } @Builder @AllArgsConstructor @Getter @NoArgsConstructor public static class Configuration { private String url; private String userName; private String password; @Builder.Default private Duration maxConnectionLifetime = Duration.ofMillis(-1); @Builder.Default private int maxConnections = 100; @Inject public Configuration(Config config) { this(); Objects.requireNonNull(config, "Config cannot be null"); url = config.getString(ServiceConfigKeys.SERVICE_DB_URL_KEY); PasswordManager passwordManager = PasswordManager.getInstance(ConfigUtils.configToProperties(config)); userName = config.getString(ServiceConfigKeys.SERVICE_DB_USERNAME); password = passwordManager.readPassword(config.getString(ServiceConfigKeys.SERVICE_DB_PASSWORD)); if(config.hasPath(ServiceConfigKeys.SERVICE_DB_MAX_CONNECTIONS)){ maxConnections = config.getInt(ServiceConfigKeys.SERVICE_DB_MAX_CONNECTIONS); } if(config.hasPath(ServiceConfigKeys.SERVICE_DB_MAX_CONNECTION_LIFETIME)){ maxConnectionLifetime = config.getDuration(ServiceConfigKeys.SERVICE_DB_MAX_CONNECTION_LIFETIME); } } } }
3,911
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/restli/GobblinServiceFlowConfigV2ResourceHandlerWithWarmStandby.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.restli; import com.google.common.base.Optional; import com.linkedin.data.transform.DataProcessingException; import com.linkedin.restli.common.ComplexResourceKey; import com.linkedin.restli.common.EmptyRecord; import com.linkedin.restli.common.HttpStatus; import com.linkedin.restli.common.PatchRequest; import com.linkedin.restli.server.CreateResponse; import com.linkedin.restli.server.UpdateResponse; import com.linkedin.restli.server.util.PatchApplier; import java.util.Properties; import javax.inject.Inject; import javax.inject.Named; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.service.FlowConfig; import org.apache.gobblin.service.FlowConfigLoggedException; import org.apache.gobblin.service.FlowConfigResourceLocalHandler; import org.apache.gobblin.service.FlowConfigV2ResourceLocalHandler; import org.apache.gobblin.service.FlowId; import org.apache.gobblin.service.modules.scheduler.GobblinServiceJobScheduler; import org.apache.gobblin.runtime.util.InjectionNames; import org.apache.helix.HelixManager; @Slf4j public class GobblinServiceFlowConfigV2ResourceHandlerWithWarmStandby extends GobblinServiceFlowConfigV2ResourceHandler { @Inject public GobblinServiceFlowConfigV2ResourceHandlerWithWarmStandby(@Named(InjectionNames.SERVICE_NAME) String serviceName, @Named(InjectionNames.FLOW_CATALOG_LOCAL_COMMIT) boolean flowCatalogLocalCommit, FlowConfigV2ResourceLocalHandler handler, Optional<HelixManager> manager, GobblinServiceJobScheduler jobScheduler, @Named(InjectionNames.FORCE_LEADER) boolean forceLeader) { super(serviceName, flowCatalogLocalCommit, handler, manager, jobScheduler, forceLeader); } @Override public UpdateResponse deleteFlowConfig(FlowId flowId, Properties header) throws FlowConfigLoggedException { return this.localHandler.deleteFlowConfig(flowId, header); } @Override public UpdateResponse partialUpdateFlowConfig(FlowId flowId, PatchRequest<FlowConfig> flowConfigPatch) throws FlowConfigLoggedException { long modifiedWatermark = System.currentTimeMillis() / 1000; FlowConfig flowConfig = getFlowConfig(flowId); try { PatchApplier.applyPatch(flowConfig, flowConfigPatch); } catch (DataProcessingException e) { throw new FlowConfigLoggedException(HttpStatus.S_400_BAD_REQUEST, "Failed to apply partial update", e); } return updateFlowConfig(flowId, flowConfig, modifiedWatermark); } @Override public UpdateResponse updateFlowConfig(FlowId flowId, FlowConfig flowConfig) throws FlowConfigLoggedException { // We have modifiedWatermark here to avoid update config happens at the same time on different hosts overwrite each other // timestamp here will be treated as largest modifiedWatermark that we can update long version = System.currentTimeMillis() / 1000; return updateFlowConfig(flowId, flowConfig, version); } public UpdateResponse updateFlowConfig(FlowId flowId, FlowConfig flowConfig, long modifiedWatermark) throws FlowConfigLoggedException { String flowName = flowId.getFlowName(); String flowGroup = flowId.getFlowGroup(); if (!flowGroup.equals(flowConfig.getId().getFlowGroup()) || !flowName.equals(flowConfig.getId().getFlowName())) { throw new FlowConfigLoggedException(HttpStatus.S_400_BAD_REQUEST, "flowName and flowGroup cannot be changed in update", null); } // We directly call localHandler to create flow config and put it in spec store //Instead of helix message, forwarding message is done by change stream of spec store return this.localHandler.updateFlowConfig(flowId, flowConfig, true, modifiedWatermark); } /** * Adding {@link FlowConfig} call {@link FlowConfigResourceLocalHandler#createFlowConfig(FlowConfig)} directly. * no matter it's active or standby, rely on the CDC stream for spec store to forward the change to other hosts * */ @Override public CreateResponse createFlowConfig(FlowConfig flowConfig) throws FlowConfigLoggedException { if (flowConfig.getProperties().containsKey(ConfigurationKeys.FLOW_EXECUTION_ID_KEY)) { throw new FlowConfigLoggedException(HttpStatus.S_400_BAD_REQUEST, String.format("%s cannot be set by the user", ConfigurationKeys.FLOW_EXECUTION_ID_KEY), null); } CreateResponse response = null; // We directly call localHandler to create flow config and put it in spec store response = this.localHandler.createFlowConfig(flowConfig, true); //Instead of helix message, forwarding message is done by change stream of spec store // Do actual work on remote node, directly return success return response == null ? new CreateResponse(new ComplexResourceKey<>(flowConfig.getId(), new EmptyRecord()), HttpStatus.S_201_CREATED) : response; } }
3,912
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/restli/GobblinServiceFlowConfigResourceHandler.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.restli; import java.io.IOException; import java.util.Collection; import java.util.Properties; import java.util.UUID; import org.apache.helix.HelixManager; import org.apache.helix.InstanceType; import com.google.common.base.Optional; import com.linkedin.data.transform.DataProcessingException; import com.linkedin.restli.common.ComplexResourceKey; import com.linkedin.restli.common.EmptyRecord; import com.linkedin.restli.common.HttpStatus; import com.linkedin.restli.common.PatchRequest; import com.linkedin.restli.server.CreateResponse; import com.linkedin.restli.server.UpdateResponse; import com.linkedin.restli.server.util.PatchApplier; import javax.inject.Inject; import javax.inject.Named; import lombok.Getter; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.runtime.api.FlowSpecSearchObject; import org.apache.gobblin.service.FlowConfig; import org.apache.gobblin.service.FlowConfigLoggedException; import org.apache.gobblin.service.FlowConfigResourceLocalHandler; import org.apache.gobblin.service.FlowConfigsResourceHandler; import org.apache.gobblin.service.FlowId; import org.apache.gobblin.service.ServiceConfigKeys; import org.apache.gobblin.service.modules.scheduler.GobblinServiceJobScheduler; import org.apache.gobblin.service.modules.utils.HelixUtils; import org.apache.gobblin.runtime.util.InjectionNames; /** * An HA (high available) aware {@link FlowConfigsResourceHandler} which consider if current node is Active or Standby. * When a Standby mode detected, it will forward the rest-li request ({@link FlowConfig}) * to the Active. Otherwise it will handle it locally. */ @Slf4j public class GobblinServiceFlowConfigResourceHandler implements FlowConfigsResourceHandler { @Getter private String serviceName; private boolean flowCatalogLocalCommit; protected FlowConfigResourceLocalHandler localHandler; private Optional<HelixManager> helixManager; private GobblinServiceJobScheduler jobScheduler; private boolean forceLeader; @Inject public GobblinServiceFlowConfigResourceHandler(@Named(InjectionNames.SERVICE_NAME) String serviceName, @Named(InjectionNames.FLOW_CATALOG_LOCAL_COMMIT) boolean flowCatalogLocalCommit, FlowConfigResourceLocalHandler handler, Optional<HelixManager> manager, GobblinServiceJobScheduler jobScheduler, @Named(InjectionNames.FORCE_LEADER) boolean forceLeader) { this.flowCatalogLocalCommit = flowCatalogLocalCommit; this.serviceName = serviceName; this.localHandler = handler; this.helixManager = manager; this.jobScheduler = jobScheduler; this.forceLeader = forceLeader; } @Override public FlowConfig getFlowConfig(FlowId flowId) throws FlowConfigLoggedException { return this.localHandler.getFlowConfig(flowId); } @Override public Collection<FlowConfig> getFlowConfig(FlowSpecSearchObject flowSpecSearchObject) throws FlowConfigLoggedException { return this.localHandler.getFlowConfig(flowSpecSearchObject); } @Override public Collection<FlowConfig> getAllFlowConfigs() { return this.localHandler.getAllFlowConfigs(); } @Override public Collection<FlowConfig> getAllFlowConfigs(int start, int count) { return this.localHandler.getAllFlowConfigs(start, count); } /** * Adding {@link FlowConfig} should check if current node is active (master). * If current node is active, call {@link FlowConfigResourceLocalHandler#createFlowConfig(FlowConfig)} directly. * If current node is standby, forward {@link ServiceConfigKeys#HELIX_FLOWSPEC_ADD} to active. The remote active will * then call {@link FlowConfigResourceLocalHandler#createFlowConfig(FlowConfig)}. * * Please refer to {@link org.apache.gobblin.service.modules.core.ControllerUserDefinedMessageHandlerFactory} for remote handling. * * For better I/O load balance, user can enable {@link GobblinServiceFlowConfigResourceHandler#flowCatalogLocalCommit}. * The {@link FlowConfig} will be then persisted to {@link org.apache.gobblin.runtime.spec_catalog.FlowCatalog} first before it is * forwarded to active node (if current node is standby) for execution. */ @Override public CreateResponse createFlowConfig(FlowConfig flowConfig) throws FlowConfigLoggedException { String flowName = flowConfig.getId().getFlowName(); String flowGroup = flowConfig.getId().getFlowGroup(); if (flowConfig.getProperties().containsKey(ConfigurationKeys.FLOW_EXECUTION_ID_KEY)) { throw new FlowConfigLoggedException(HttpStatus.S_400_BAD_REQUEST, String.format("%s cannot be set by the user", ConfigurationKeys.FLOW_EXECUTION_ID_KEY), null); } checkHelixConnection(ServiceConfigKeys.HELIX_FLOWSPEC_ADD, flowName, flowGroup); if (forceLeader) { HelixUtils.throwErrorIfNotLeader(helixManager); } try { if (!jobScheduler.isActive() && helixManager.isPresent()) { CreateResponse response = null; if (this.flowCatalogLocalCommit) { // We will handle FS I/O locally for load balance before forwarding to remote node. response = this.localHandler.createFlowConfig(flowConfig, true); } if (!flowConfig.hasExplain() || !flowConfig.isExplain()) { //Forward the message to master only if it is not an "explain" request. forwardMessage(ServiceConfigKeys.HELIX_FLOWSPEC_ADD, FlowConfigUtils.serializeFlowConfig(flowConfig), flowName, flowGroup); } // Do actual work on remote node, directly return success return response == null ? new CreateResponse(new ComplexResourceKey<>(flowConfig.getId(), new EmptyRecord()), HttpStatus.S_201_CREATED) : response; } else { return this.localHandler.createFlowConfig(flowConfig); } } catch (IOException e) { throw new FlowConfigLoggedException(HttpStatus.S_500_INTERNAL_SERVER_ERROR, "Cannot create flowConfig [flowName=" + flowName + " flowGroup=" + flowGroup + "]", e); } } /** * Updating {@link FlowConfig} should check if current node is active (master). * If current node is active, call {@link FlowConfigResourceLocalHandler#updateFlowConfig(FlowId, FlowConfig)} directly. * If current node is standby, forward {@link ServiceConfigKeys#HELIX_FLOWSPEC_UPDATE} to active. The remote active will * then call {@link FlowConfigResourceLocalHandler#updateFlowConfig(FlowId, FlowConfig)}. * * Please refer to {@link org.apache.gobblin.service.modules.core.ControllerUserDefinedMessageHandlerFactory} for remote handling. * * For better I/O load balance, user can enable {@link GobblinServiceFlowConfigResourceHandler#flowCatalogLocalCommit}. * The {@link FlowConfig} will be then persisted to {@link org.apache.gobblin.runtime.spec_catalog.FlowCatalog} first before it is * forwarded to active node (if current node is standby) for execution. */ @Override public UpdateResponse updateFlowConfig(FlowId flowId, FlowConfig flowConfig) throws FlowConfigLoggedException { String flowName = flowId.getFlowName(); String flowGroup = flowId.getFlowGroup(); if (!flowGroup.equals(flowConfig.getId().getFlowGroup()) || !flowName.equals(flowConfig.getId().getFlowName())) { throw new FlowConfigLoggedException(HttpStatus.S_400_BAD_REQUEST, "flowName and flowGroup cannot be changed in update", null); } checkHelixConnection(ServiceConfigKeys.HELIX_FLOWSPEC_UPDATE, flowName, flowGroup); if (forceLeader) { HelixUtils.throwErrorIfNotLeader(helixManager); } try { if (!jobScheduler.isActive() && helixManager.isPresent()) { if (this.flowCatalogLocalCommit) { // We will handle FS I/O locally for load balance before forwarding to remote node. this.localHandler.updateFlowConfig(flowId, flowConfig, false); } forwardMessage(ServiceConfigKeys.HELIX_FLOWSPEC_UPDATE, FlowConfigUtils.serializeFlowConfig(flowConfig), flowName, flowGroup); // Do actual work on remote node, directly return success log.info("Forwarding update flowConfig [flowName=" + flowName + " flowGroup=" + flowGroup + "]"); return new UpdateResponse(HttpStatus.S_200_OK); } else { return this.localHandler.updateFlowConfig(flowId, flowConfig); } } catch (IOException e) { throw new FlowConfigLoggedException(HttpStatus.S_500_INTERNAL_SERVER_ERROR, "Cannot update flowConfig [flowName=" + flowName + " flowGroup=" + flowGroup + "]", e); } } @Override public UpdateResponse partialUpdateFlowConfig(FlowId flowId, PatchRequest<FlowConfig> flowConfigPatch) { FlowConfig flowConfig = getFlowConfig(flowId); try { PatchApplier.applyPatch(flowConfig, flowConfigPatch); } catch (DataProcessingException e) { throw new FlowConfigLoggedException(HttpStatus.S_400_BAD_REQUEST, "Failed to apply partial update", e); } return updateFlowConfig(flowId, flowConfig); } /** * Deleting {@link FlowConfig} should check if current node is active (master). * If current node is active, call {@link FlowConfigResourceLocalHandler#deleteFlowConfig(FlowId, Properties)} directly. * If current node is standby, forward {@link ServiceConfigKeys#HELIX_FLOWSPEC_REMOVE} to active. The remote active will * then call {@link FlowConfigResourceLocalHandler#deleteFlowConfig(FlowId, Properties)}. * * Please refer to {@link org.apache.gobblin.service.modules.core.ControllerUserDefinedMessageHandlerFactory} for remote handling. * * For better I/O load balance, user can enable {@link GobblinServiceFlowConfigResourceHandler#flowCatalogLocalCommit}. * The {@link FlowConfig} will be then persisted to {@link org.apache.gobblin.runtime.spec_catalog.FlowCatalog} first before it is * forwarded to active node (if current node is standby) for execution. */ @Override public UpdateResponse deleteFlowConfig(FlowId flowId, Properties header) throws FlowConfigLoggedException { String flowName = flowId.getFlowName(); String flowGroup = flowId.getFlowGroup(); checkHelixConnection(ServiceConfigKeys.HELIX_FLOWSPEC_REMOVE, flowName, flowGroup); if (forceLeader) { HelixUtils.throwErrorIfNotLeader(helixManager); } try { if (!jobScheduler.isActive() && helixManager.isPresent()) { if (this.flowCatalogLocalCommit) { // We will handle FS I/O locally for load balance before forwarding to remote node. this.localHandler.deleteFlowConfig(flowId, header, false); } forwardMessage(ServiceConfigKeys.HELIX_FLOWSPEC_REMOVE, FlowConfigUtils.serializeFlowId(flowId), flowName, flowGroup); return new UpdateResponse(HttpStatus.S_200_OK); } else { return this.localHandler.deleteFlowConfig(flowId, header); } } catch (IOException e) { throw new FlowConfigLoggedException(HttpStatus.S_500_INTERNAL_SERVER_ERROR, "Cannot delete flowConfig [flowName=" + flowName + " flowGroup=" + flowGroup + "]", e); } } private void checkHelixConnection(String opr, String flowName, String flowGroup) throws FlowConfigLoggedException { if (this.helixManager.isPresent() && !this.helixManager.get().isConnected()) { // Specs in store will be notified when Scheduler is added as listener to FlowCatalog, so ignore // .. Specs if in cluster mode and Helix is not yet initialized log.warn("System not yet initialized. Skipping operation " + opr); throw new FlowConfigLoggedException(HttpStatus.S_500_INTERNAL_SERVER_ERROR, "System not yet initialized. Skipping " + opr + " flowConfig [flowName=" + flowName + " flowGroup=" + flowGroup + "]"); } } private void forwardMessage(String msgSubType, String val, String flowName, String flowGroup) { HelixUtils.sendUserDefinedMessage(msgSubType, val, UUID.randomUUID().toString(), InstanceType.CONTROLLER, helixManager.get(), log); log.info("{} Forwarding {} flowConfig [flowName={} flowGroup={}", serviceName, msgSubType, flowName, flowGroup + "]"); } }
3,913
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/restli/FlowConfigUtils.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.restli; import java.io.IOException; import java.util.List; import java.util.Map; import java.util.Properties; import java.util.stream.Collectors; import com.google.common.collect.Maps; import com.linkedin.data.template.StringMap; import com.typesafe.config.Config; import org.apache.gobblin.service.FlowConfig; import org.apache.gobblin.service.FlowId; import org.apache.gobblin.service.Schedule; import org.apache.gobblin.util.ConfigUtils; import org.apache.gobblin.util.PropertiesUtils; public class FlowConfigUtils { private static final String FLOWCONFIG = "fc"; private static final String FLOWCONFIG_ID = FLOWCONFIG + '-' + "id"; private static final String FLOWCONFIG_ID_NAME = FLOWCONFIG_ID + '-' + "name"; private static final String FLOWCONFIG_ID_GROUP = FLOWCONFIG_ID + '-' + "group"; private static final String FLOWCONFIG_SCHEDULE = FLOWCONFIG + '-' + "sch"; private static final String FLOWCONFIG_SCHEDULE_CRON = FLOWCONFIG_SCHEDULE + '-' + "cron"; private static final String FLOWCONFIG_SCHEDULE_RUN_IMMEDIATELY = FLOWCONFIG_SCHEDULE + '-' + "runImmediately"; private static final String FLOWCONFIG_TEMPLATEURIS = FLOWCONFIG + '-' + "templateUris"; public static String serializeFlowId(FlowId id) throws IOException { Properties properties = new Properties(); properties.setProperty(FLOWCONFIG_ID_NAME, id.getFlowName()); properties.setProperty(FLOWCONFIG_ID_GROUP, id.getFlowGroup()); return PropertiesUtils.serialize(properties); } public static FlowId deserializeFlowId(String serialized) throws IOException { Properties properties = PropertiesUtils.deserialize(serialized); FlowId id = new FlowId(); id.setFlowName(properties.getProperty(FLOWCONFIG_ID_NAME)); id.setFlowGroup(properties.getProperty(FLOWCONFIG_ID_GROUP)); return id; } public static String serializeFlowConfig(FlowConfig flowConfig) throws IOException { Properties properties = new Properties(); properties.putAll(flowConfig.getProperties()); properties.setProperty(FLOWCONFIG_ID_NAME, flowConfig.getId().getFlowName()); properties.setProperty(FLOWCONFIG_ID_GROUP, flowConfig.getId().getFlowGroup()); if (flowConfig.hasSchedule()) { properties.setProperty(FLOWCONFIG_SCHEDULE_CRON, flowConfig.getSchedule().getCronSchedule()); properties.setProperty(FLOWCONFIG_SCHEDULE_RUN_IMMEDIATELY, Boolean.toString(flowConfig.getSchedule().isRunImmediately())); } if (flowConfig.hasTemplateUris()) { properties.setProperty(FLOWCONFIG_TEMPLATEURIS, flowConfig.getTemplateUris()); } return PropertiesUtils.serialize(properties); } public static FlowConfig deserializeFlowConfig(String serialized) throws IOException { Properties properties = PropertiesUtils.deserialize(serialized); FlowConfig flowConfig = new FlowConfig().setId(new FlowId() .setFlowName(properties.getProperty(FLOWCONFIG_ID_NAME)) .setFlowGroup(properties.getProperty(FLOWCONFIG_ID_GROUP))); if (properties.containsKey(FLOWCONFIG_SCHEDULE_CRON)) { flowConfig.setSchedule(new Schedule() .setCronSchedule(properties.getProperty(FLOWCONFIG_SCHEDULE_CRON)) .setRunImmediately(Boolean.valueOf(properties.getProperty(FLOWCONFIG_SCHEDULE_RUN_IMMEDIATELY)))); } if (properties.containsKey(FLOWCONFIG_TEMPLATEURIS)) { flowConfig.setTemplateUris(properties.getProperty(FLOWCONFIG_TEMPLATEURIS)); } properties.remove(FLOWCONFIG_ID_NAME); properties.remove(FLOWCONFIG_ID_GROUP); properties.remove(FLOWCONFIG_SCHEDULE_CRON); properties.remove(FLOWCONFIG_SCHEDULE_RUN_IMMEDIATELY); properties.remove(FLOWCONFIG_TEMPLATEURIS); flowConfig.setProperties(new StringMap(Maps.fromProperties(properties))); return flowConfig; } public static List<String> getDataNodes(Config flowConfig, String identifierKey, Map<String, String> dataNodeAliasMap) { List<String> dataNodes = ConfigUtils.getStringList(flowConfig, identifierKey); return dataNodes.stream().map(dataNode -> dataNodeAliasMap.getOrDefault(dataNode, dataNode)).collect(Collectors.toList()); } public static String getDataNode(Config flowConfig, String identifierKey, Map<String, String> dataNodeAliasMap) { String dataNode = ConfigUtils.getString(flowConfig, identifierKey, ""); return dataNodeAliasMap.getOrDefault(dataNode, dataNode); } }
3,914
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/restli/GobblinServiceFlowConfigV2ResourceHandler.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.restli; import org.apache.helix.HelixManager; import com.google.common.base.Optional; import javax.inject.Inject; import javax.inject.Named; import org.apache.gobblin.service.FlowConfigV2ResourceLocalHandler; import org.apache.gobblin.service.FlowConfigsV2ResourceHandler; import org.apache.gobblin.service.modules.scheduler.GobblinServiceJobScheduler; import org.apache.gobblin.runtime.util.InjectionNames; public class GobblinServiceFlowConfigV2ResourceHandler extends GobblinServiceFlowConfigResourceHandler implements FlowConfigsV2ResourceHandler { @Inject public GobblinServiceFlowConfigV2ResourceHandler(@Named(InjectionNames.SERVICE_NAME) String serviceName, @Named(InjectionNames.FLOW_CATALOG_LOCAL_COMMIT) boolean flowCatalogLocalCommit, FlowConfigV2ResourceLocalHandler handler, Optional<HelixManager> manager, GobblinServiceJobScheduler jobScheduler, @Named(InjectionNames.FORCE_LEADER) boolean forceLeader) { super(serviceName, flowCatalogLocalCommit, handler, manager, jobScheduler, forceLeader); } }
3,915
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/restli/GobblinServiceFlowExecutionResourceHandlerWithWarmStandby.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.restli; import com.google.common.base.Optional; import com.google.common.eventbus.EventBus; import com.google.inject.Inject; import com.linkedin.restli.common.ComplexResourceKey; import com.linkedin.restli.common.EmptyRecord; import com.linkedin.restli.common.HttpStatus; import com.linkedin.restli.server.RestLiServiceException; import com.linkedin.restli.server.UpdateResponse; import java.io.IOException; import java.sql.SQLException; import javax.inject.Named; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.runtime.api.DagActionStore; import org.apache.gobblin.runtime.util.InjectionNames; import org.apache.gobblin.service.FlowExecutionResourceLocalHandler; import org.apache.gobblin.service.modules.core.GobblinServiceManager; import org.apache.helix.HelixManager; @Slf4j public class GobblinServiceFlowExecutionResourceHandlerWithWarmStandby extends GobblinServiceFlowExecutionResourceHandler{ private DagActionStore dagActionStore; @Inject public GobblinServiceFlowExecutionResourceHandlerWithWarmStandby(FlowExecutionResourceLocalHandler handler, @Named(GobblinServiceManager.SERVICE_EVENT_BUS_NAME) EventBus eventBus, Optional<HelixManager> manager, @Named(InjectionNames.FORCE_LEADER) boolean forceLeader, DagActionStore dagActionStore) { super(handler, eventBus, manager, forceLeader); this.dagActionStore = dagActionStore; } @Override public void resume(ComplexResourceKey<org.apache.gobblin.service.FlowStatusId, EmptyRecord> key) { String flowGroup = key.getKey().getFlowGroup(); String flowName = key.getKey().getFlowName(); Long flowExecutionId = key.getKey().getFlowExecutionId(); try { // If an existing resume request is still pending then do not accept this request if (this.dagActionStore.exists(flowGroup, flowName, flowExecutionId.toString(), DagActionStore.FlowActionType.RESUME)) { this.prepareError("There is already a pending RESUME action for this flow. Please wait to resubmit and wait " + "for action to be completed.", HttpStatus.S_409_CONFLICT); return; } this.dagActionStore.addDagAction(flowGroup, flowName, flowExecutionId.toString(), DagActionStore.FlowActionType.RESUME); } catch (IOException | SQLException e) { log.warn( String.format("Failed to add execution resume action for flow %s %s %s to dag action store due to", flowGroup, flowName, flowExecutionId), e); this.prepareError(e.getMessage(), HttpStatus.S_500_INTERNAL_SERVER_ERROR); } } private void prepareError(String exceptionMessage, HttpStatus errorType) { if (errorType == HttpStatus.S_409_CONFLICT) { throw new RestLiServiceException(HttpStatus.S_409_CONFLICT, exceptionMessage); } else if (errorType == HttpStatus.S_400_BAD_REQUEST) { throw new RestLiServiceException(HttpStatus.S_400_BAD_REQUEST); } throw new RestLiServiceException(HttpStatus.S_500_INTERNAL_SERVER_ERROR, exceptionMessage); } @Override public UpdateResponse delete(ComplexResourceKey<org.apache.gobblin.service.FlowStatusId, EmptyRecord> key) { String flowGroup = key.getKey().getFlowGroup(); String flowName = key.getKey().getFlowName(); Long flowExecutionId = key.getKey().getFlowExecutionId(); try { // If an existing kill request is still pending then do not accept this request if (this.dagActionStore.exists(flowGroup, flowName, flowExecutionId.toString(), DagActionStore.FlowActionType.KILL)) { this.prepareError("There is already a pending KILL action for this flow. Please wait to resubmit and wait " + "for action to be completed.", HttpStatus.S_400_BAD_REQUEST); } this.dagActionStore.addDagAction(flowGroup, flowName, flowExecutionId.toString(), DagActionStore.FlowActionType.KILL); return new UpdateResponse(HttpStatus.S_200_OK); } catch (IOException | SQLException e) { this.prepareError(String.format("Failed to add execution delete action for flow %s %s %s to dag action store due to", flowGroup, flowName, flowExecutionId), HttpStatus.S_500_INTERNAL_SERVER_ERROR); return new UpdateResponse(HttpStatus.S_500_INTERNAL_SERVER_ERROR); } } }
3,916
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/restli/GobblinServiceFlowExecutionResourceHandler.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.restli; import java.util.List; import org.apache.helix.HelixManager; import com.google.common.base.Optional; import com.google.common.eventbus.EventBus; import com.linkedin.restli.common.ComplexResourceKey; import com.linkedin.restli.common.EmptyRecord; import com.linkedin.restli.common.HttpStatus; import com.linkedin.restli.server.PagingContext; import com.linkedin.restli.server.UpdateResponse; import javax.inject.Inject; import javax.inject.Named; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.service.FlowExecution; import org.apache.gobblin.service.FlowExecutionResourceHandler; import org.apache.gobblin.service.FlowExecutionResourceLocalHandler; import org.apache.gobblin.service.FlowId; import org.apache.gobblin.service.FlowStatusId; import org.apache.gobblin.service.modules.core.GobblinServiceManager; import org.apache.gobblin.service.modules.utils.HelixUtils; import org.apache.gobblin.runtime.util.InjectionNames; import org.apache.gobblin.service.monitoring.KillFlowEvent; import org.apache.gobblin.service.monitoring.ResumeFlowEvent; /** * {@link FlowExecutionResourceHandler} that calls underlying resource handler, but does extra work that requires objects * like the {@link HelixManager}. For now, that is just checking leadership and sending the kill through the eventBus * for the delete method. */ @Slf4j public class GobblinServiceFlowExecutionResourceHandler implements FlowExecutionResourceHandler { private FlowExecutionResourceLocalHandler localHandler; private EventBus eventBus; private Optional<HelixManager> helixManager; private boolean forceLeader; @Inject public GobblinServiceFlowExecutionResourceHandler(FlowExecutionResourceLocalHandler handler, @Named(GobblinServiceManager.SERVICE_EVENT_BUS_NAME) EventBus eventBus, Optional<HelixManager> manager, @Named(InjectionNames.FORCE_LEADER) boolean forceLeader) { this.localHandler = handler; this.eventBus = eventBus; this.helixManager = manager; this.forceLeader = forceLeader; } @Override public FlowExecution get(ComplexResourceKey<FlowStatusId, EmptyRecord> key) { return this.localHandler.get(key); } @Override public List<FlowExecution> getLatestFlowExecution(PagingContext context, FlowId flowId, Integer count, String tag, String executionStatus, Boolean includeIssues) { return this.localHandler.getLatestFlowExecution(context, flowId, count, tag, executionStatus, includeIssues); } @Override public List<FlowExecution> getLatestFlowGroupExecutions(PagingContext context, String flowGroup, Integer countPerFlow, String tag, Boolean includeIssues) { return this.localHandler.getLatestFlowGroupExecutions(context, flowGroup, countPerFlow, tag, includeIssues); } @Override public void resume(ComplexResourceKey<FlowStatusId, EmptyRecord> key) { String flowGroup = key.getKey().getFlowGroup(); String flowName = key.getKey().getFlowName(); Long flowExecutionId = key.getKey().getFlowExecutionId(); if (this.forceLeader) { HelixUtils.throwErrorIfNotLeader(this.helixManager); } this.eventBus.post(new ResumeFlowEvent(flowGroup, flowName, flowExecutionId)); } @Override public UpdateResponse delete(ComplexResourceKey<FlowStatusId, EmptyRecord> key) { String flowGroup = key.getKey().getFlowGroup(); String flowName = key.getKey().getFlowName(); Long flowExecutionId = key.getKey().getFlowExecutionId(); if (this.forceLeader) { HelixUtils.throwErrorIfNotLeader(this.helixManager); } this.eventBus.post(new KillFlowEvent(flowGroup, flowName, flowExecutionId)); return new UpdateResponse(HttpStatus.S_200_OK); } }
3,917
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/flow/IdentityFlowToJobSpecCompiler.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.flow; import java.util.ArrayList; import java.util.List; import java.util.Map; import java.util.concurrent.ExecutionException; import java.util.concurrent.TimeUnit; import org.slf4j.Logger; import com.google.common.base.Optional; import com.google.common.base.Preconditions; import com.typesafe.config.Config; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.runtime.api.FlowSpec; import org.apache.gobblin.runtime.api.JobSpec; import org.apache.gobblin.runtime.api.ServiceNode; import org.apache.gobblin.runtime.api.Spec; import org.apache.gobblin.runtime.api.SpecExecutor; import org.apache.gobblin.runtime.api.TopologySpec; import org.apache.gobblin.service.ServiceConfigKeys; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.service.modules.spec.JobExecutionPlanDagFactory; /*** * Take in a logical {@link Spec} ie flow and compile corresponding materialized job {@link Spec} * and its mapping to {@link SpecExecutor}. */ @Alpha public class IdentityFlowToJobSpecCompiler extends BaseFlowToJobSpecCompiler { public IdentityFlowToJobSpecCompiler(Config config) { super(config, true); } public IdentityFlowToJobSpecCompiler(Config config, boolean instrumentationEnabled) { super(config, Optional.<Logger>absent(), instrumentationEnabled); } public IdentityFlowToJobSpecCompiler(Config config, Optional<Logger> log) { super(config, log, true); } public IdentityFlowToJobSpecCompiler(Config config, Optional<Logger> log, boolean instrumentationEnabled) { super(config, log, instrumentationEnabled); } @Override public Dag<JobExecutionPlan> compileFlow(Spec spec) { Preconditions.checkNotNull(spec); Preconditions.checkArgument(spec instanceof FlowSpec, "IdentityFlowToJobSpecCompiler only converts FlowSpec to JobSpec"); long startTime = System.nanoTime(); FlowSpec flowSpec = (FlowSpec) spec; String source = flowSpec.getConfig().getString(ServiceConfigKeys.FLOW_SOURCE_IDENTIFIER_KEY); String destination = flowSpec.getConfig().getString(ServiceConfigKeys.FLOW_DESTINATION_IDENTIFIER_KEY); log.info(String.format("Compiling flow for source: %s and destination: %s", source, destination)); JobSpec jobSpec = jobSpecGenerator(flowSpec); Instrumented.markMeter(this.flowCompilationSuccessFulMeter); Instrumented.updateTimer(this.flowCompilationTimer, System.nanoTime() - startTime, TimeUnit.NANOSECONDS); List<JobExecutionPlan> jobExecutionPlans; try { jobExecutionPlans = getJobExecutionPlans(source, destination, jobSpec); } catch (InterruptedException | ExecutionException e) { Instrumented.markMeter(this.flowCompilationFailedMeter); throw new RuntimeException("Cannot determine topology capabilities", e); } return new JobExecutionPlanDagFactory().createDag(jobExecutionPlans); } private List<JobExecutionPlan> getJobExecutionPlans(String source, String destination, JobSpec jobSpec) throws ExecutionException, InterruptedException { List<JobExecutionPlan> jobExecutionPlans = new ArrayList<>(); for (TopologySpec topologySpec : topologySpecMap.values()) { Map<ServiceNode, ServiceNode> capabilities = topologySpec.getSpecExecutor().getCapabilities().get(); for (Map.Entry<ServiceNode, ServiceNode> capability : capabilities.entrySet()) { log.info(String.format("Evaluating current JobSpec: %s against TopologySpec: %s with " + "capability of source: %s and destination: %s ", jobSpec.getUri(), topologySpec.getUri(), capability.getKey(), capability.getValue())); if (source.equals(capability.getKey().getNodeName()) && destination .equals(capability.getValue().getNodeName())) { JobExecutionPlan jobExecutionPlan = new JobExecutionPlan(jobSpec, topologySpec.getSpecExecutor()); log.info(String .format("Current JobSpec: %s is executable on TopologySpec: %s. Added TopologySpec as candidate.", jobSpec.getUri(), topologySpec.getUri())); log.info("Since we found a candidate executor, we will not try to compute more. " + "(Intended limitation for IdentityFlowToJobSpecCompiler)"); jobExecutionPlans.add(jobExecutionPlan); return jobExecutionPlans; } } } return jobExecutionPlans; } }
3,918
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/flow/MockedSpecCompiler.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.flow; import java.util.ArrayList; import java.util.List; import java.util.Properties; import com.typesafe.config.Config; import com.typesafe.config.ConfigFactory; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.runtime.api.FlowSpec; import org.apache.gobblin.runtime.api.JobSpec; import org.apache.gobblin.runtime.api.Spec; import org.apache.gobblin.runtime.spec_executorInstance.InMemorySpecExecutor; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.service.modules.spec.JobExecutionPlanDagFactory; import org.apache.gobblin.util.ConfigUtils; /** * This mocked SpecCompiler class creates 3 dummy job specs to emulate flow spec compiler. * It can also be used to compile in a certain way or not to compile at all to write negative test cases. * It uses {@link InMemorySpecExecutor} for these dummy specs. */ public class MockedSpecCompiler extends IdentityFlowToJobSpecCompiler { private static final int NUMBER_OF_JOBS = 3; public static final String UNCOMPILABLE_FLOW = "uncompilableFlow"; public MockedSpecCompiler(Config config) { super(config); } @Override public Dag<JobExecutionPlan> compileFlow(Spec spec) { String flowName = (String) ((FlowSpec) spec).getConfigAsProperties().get(ConfigurationKeys.FLOW_NAME_KEY); if (flowName.equalsIgnoreCase(UNCOMPILABLE_FLOW)) { return null; } List<JobExecutionPlan> jobExecutionPlans = new ArrayList<>(); long flowExecutionId = System.currentTimeMillis(); int i = 0; while(i++ < NUMBER_OF_JOBS) { String specUri = "/foo/bar/spec/" + i; Properties properties = new Properties(); properties.put(ConfigurationKeys.FLOW_NAME_KEY, flowName); properties.put(ConfigurationKeys.FLOW_GROUP_KEY, ((FlowSpec)spec).getConfigAsProperties().get(ConfigurationKeys.FLOW_GROUP_KEY)); properties.put(ConfigurationKeys.JOB_NAME_KEY, ((FlowSpec)spec).getConfigAsProperties().get(ConfigurationKeys.FLOW_NAME_KEY) + "_" + i); properties.put(ConfigurationKeys.JOB_GROUP_KEY, ((FlowSpec)spec).getConfigAsProperties().get(ConfigurationKeys.FLOW_GROUP_KEY) + "_" + i); properties.put(ConfigurationKeys.FLOW_EXECUTION_ID_KEY, flowExecutionId); JobSpec jobSpec = JobSpec.builder(specUri) .withConfig(ConfigUtils.propertiesToConfig(properties)) .withVersion("1") .withDescription("Spec Description") .build(); jobExecutionPlans.add(new JobExecutionPlan(jobSpec, new InMemorySpecExecutor(ConfigFactory.empty()))); } return new JobExecutionPlanDagFactory().createDag(jobExecutionPlans); } }
3,919
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/flow/FlowEdgeContext.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.flow; import com.typesafe.config.Config; import lombok.AllArgsConstructor; import lombok.EqualsAndHashCode; import lombok.Getter; import org.apache.gobblin.runtime.api.SpecExecutor; import org.apache.gobblin.service.modules.dataset.DatasetDescriptor; import org.apache.gobblin.service.modules.flowgraph.FlowEdge; /** * A helper class used to maintain additional context associated with each {@link FlowEdge} during path * computation while the edge is explored for its eligibility. The additional context includes the input * {@link DatasetDescriptor} of this edge which is compatible with the previous {@link FlowEdge}'s output * {@link DatasetDescriptor} (where "previous" means the immediately preceding {@link FlowEdge} visited before * the current {@link FlowEdge}), and the corresponding output dataset descriptor of the current {@link FlowEdge}. */ @AllArgsConstructor @EqualsAndHashCode(exclude = {"mergedConfig", "specExecutor"}) @Getter public class FlowEdgeContext { private FlowEdge edge; private DatasetDescriptor inputDatasetDescriptor; private DatasetDescriptor outputDatasetDescriptor; private Config mergedConfig; private SpecExecutor specExecutor; @Override public String toString() { return edge == null ? "Null" : edge.toString(); } }
3,920
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/flow/BaseFlowToJobSpecCompiler.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.flow; import java.io.IOException; import java.net.URI; import java.util.Collections; import java.util.List; import java.util.Map; import java.util.Properties; import org.apache.commons.lang3.StringUtils; import org.apache.gobblin.runtime.spec_catalog.FlowCatalog; import org.apache.gobblin.service.modules.orchestration.UserQuotaManager; import org.apache.gobblin.util.reflection.GobblinConstructorUtils; import org.quartz.CronExpression; import org.slf4j.Logger; import org.slf4j.LoggerFactory; import com.codahale.metrics.Meter; import com.codahale.metrics.Timer; import com.google.common.base.Optional; import com.google.common.collect.Maps; import com.typesafe.config.Config; import com.typesafe.config.ConfigValueFactory; import javax.annotation.Nonnull; import lombok.Getter; import lombok.Setter; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.configuration.State; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.metrics.MetricContext; import org.apache.gobblin.metrics.Tag; import org.apache.gobblin.runtime.api.FlowSpec; import org.apache.gobblin.runtime.api.JobSpec; import org.apache.gobblin.runtime.api.JobTemplate; import org.apache.gobblin.runtime.api.Spec; import org.apache.gobblin.runtime.api.SpecNotFoundException; import org.apache.gobblin.runtime.api.TopologySpec; import org.apache.gobblin.runtime.job_catalog.FSJobCatalog; import org.apache.gobblin.runtime.job_spec.ResolvedJobSpec; import org.apache.gobblin.runtime.spec_catalog.AddSpecResponse; import org.apache.gobblin.service.ServiceConfigKeys; import org.apache.gobblin.metrics.ServiceMetricNames; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.util.ConfigUtils; import org.apache.gobblin.util.PropertiesUtils; // Provide base implementation for constructing multi-hops route. @Alpha public abstract class BaseFlowToJobSpecCompiler implements SpecCompiler { // Since {@link SpecCompiler} is an {@link SpecCatalogListener}, it is expected that any Spec change should be reflected // to these data structures. @Getter @Setter protected final Map<URI, TopologySpec> topologySpecMap; protected final Config config; protected final Logger log; protected final Optional<FSJobCatalog> templateCatalog; protected final MetricContext metricContext; @Getter protected Optional<Meter> flowCompilationSuccessFulMeter; @Getter protected Optional<Meter> flowCompilationFailedMeter; @Getter protected Optional<Timer> flowCompilationTimer; @Getter protected Optional<Timer> dataAuthorizationTimer; @Getter @Setter protected boolean active; private boolean warmStandbyEnabled; private Optional<UserQuotaManager> userQuotaManager; public BaseFlowToJobSpecCompiler(Config config){ this(config,true); } public BaseFlowToJobSpecCompiler(Config config, boolean instrumentationEnabled){ this(config, Optional.<Logger>absent(), true); } public BaseFlowToJobSpecCompiler(Config config, Optional<Logger> log){ this(config, log,true); } public BaseFlowToJobSpecCompiler(Config config, Optional<Logger> log, boolean instrumentationEnabled){ this.log = log.isPresent() ? log.get() : LoggerFactory.getLogger(getClass()); if (instrumentationEnabled) { this.metricContext = Instrumented.getMetricContext(ConfigUtils.configToState(config), IdentityFlowToJobSpecCompiler.class); this.flowCompilationSuccessFulMeter = Optional.of(this.metricContext.meter(ServiceMetricNames.FLOW_COMPILATION_SUCCESSFUL_METER)); this.flowCompilationFailedMeter = Optional.of(this.metricContext.meter(ServiceMetricNames.FLOW_COMPILATION_FAILED_METER)); this.flowCompilationTimer = Optional.<Timer>of(this.metricContext.timer(ServiceMetricNames.FLOW_COMPILATION_TIMER)); this.dataAuthorizationTimer = Optional.<Timer>of(this.metricContext.timer(ServiceMetricNames.DATA_AUTHORIZATION_TIMER)); } else { this.metricContext = null; this.flowCompilationSuccessFulMeter = Optional.absent(); this.flowCompilationFailedMeter = Optional.absent(); this.flowCompilationTimer = Optional.absent(); this.dataAuthorizationTimer = Optional.absent(); } this.warmStandbyEnabled = ConfigUtils.getBoolean(config, ServiceConfigKeys.GOBBLIN_SERVICE_WARM_STANDBY_ENABLED_KEY, false); if (this.warmStandbyEnabled) { userQuotaManager = Optional.of(GobblinConstructorUtils.invokeConstructor(UserQuotaManager.class, ConfigUtils.getString(config, ServiceConfigKeys.QUOTA_MANAGER_CLASS, ServiceConfigKeys.DEFAULT_QUOTA_MANAGER), config)); } else { userQuotaManager = Optional.absent(); } this.topologySpecMap = Maps.newConcurrentMap(); this.config = config; /*** * ETL-5996 * For multi-tenancy, the following needs to be added: * 1. Change singular templateCatalog to Map<URI, JobCatalogWithTemplates> to support multiple templateCatalogs * 2. Pick templateCatalog from JobCatalogWithTemplates based on URI, and try to resolve JobSpec using that */ try { if (this.config.hasPath(ServiceConfigKeys.TEMPLATE_CATALOGS_FULLY_QUALIFIED_PATH_KEY) && StringUtils.isNotBlank(this.config.getString(ServiceConfigKeys.TEMPLATE_CATALOGS_FULLY_QUALIFIED_PATH_KEY))) { Config templateCatalogCfg = config .withValue(ConfigurationKeys.JOB_CONFIG_FILE_GENERAL_PATH_KEY, this.config.getValue(ServiceConfigKeys.TEMPLATE_CATALOGS_FULLY_QUALIFIED_PATH_KEY)); this.templateCatalog = Optional.of(new FSJobCatalog(templateCatalogCfg)); } else { this.templateCatalog = Optional.absent(); } } catch (IOException e) { throw new RuntimeException("Could not initialize FlowCompiler because of " + "TemplateCatalog initialization failure", e); } } @Override public void awaitHealthy() throws InterruptedException { //Do nothing return; } private synchronized AddSpecResponse onAddTopologySpec(TopologySpec spec) { log.info("Loading topology {}", spec.toLongString()); for (Map.Entry entry : spec.getConfigAsProperties().entrySet()) { log.info("topo: {} --> {}", entry.getKey(), entry.getValue()); } topologySpecMap.put(spec.getUri(), spec); return new AddSpecResponse(null); } private AddSpecResponse onAddFlowSpec(FlowSpec flowSpec) { Properties flowSpecProperties = flowSpec.getConfigAsProperties(); if (topologySpecMap.containsKey(flowSpec.getUri())) { log.error("flow spec URI: {} is the same as one of the spec executors uris, ignore the flow", flowSpec.getUri()); flowSpec.getCompilationErrors().add(new FlowSpec.CompilationError(0, "invalid flow spec uri " + flowSpec.getUri() + " because it is the same as one of the spec executors uri")); return null; } if (flowSpecProperties.containsKey(ConfigurationKeys.JOB_SCHEDULE_KEY) && StringUtils.isNotBlank( flowSpecProperties.getProperty(ConfigurationKeys.JOB_SCHEDULE_KEY))) { try { new CronExpression(flowSpecProperties.getProperty(ConfigurationKeys.JOB_SCHEDULE_KEY)); } catch (Exception e) { log.error("invalid cron schedule: {}", flowSpecProperties.getProperty(ConfigurationKeys.JOB_SCHEDULE_KEY), e); flowSpec.getCompilationErrors().add(new FlowSpec.CompilationError(0, "invalid cron schedule: " + flowSpecProperties.getProperty(ConfigurationKeys.JOB_SCHEDULE_KEY) + e.getMessage())); return null; } } String response = null; // always try to compile the flow to verify if it is compilable Dag<JobExecutionPlan> dag = this.compileFlow(flowSpec); // If dag is null then a compilation error has occurred if (dag != null && !dag.isEmpty()) { response = dag.toString(); } if (FlowCatalog.isCompileSuccessful(response) && this.userQuotaManager.isPresent() && !flowSpec.isExplain() && !flowSpec.getConfigAsProperties().containsKey(ConfigurationKeys.JOB_SCHEDULE_KEY)) { try { // We only check quota for adhoc flow, since we don't have the execution id for run-immediately flow userQuotaManager.get().checkQuota(dag.getStartNodes()); } catch (IOException e) { throw new RuntimeException(e); } } return new AddSpecResponse<>(response); } @Override public AddSpecResponse onAddSpec(Spec addedSpec) { if (addedSpec instanceof FlowSpec) { return onAddFlowSpec((FlowSpec) addedSpec); } else if (addedSpec instanceof TopologySpec) { return onAddTopologySpec( (TopologySpec) addedSpec); } return new AddSpecResponse(null); } public void onDeleteSpec(URI deletedSpecURI, String deletedSpecVersion) { onDeleteSpec(deletedSpecURI, deletedSpecVersion, new Properties()); } @Override public synchronized void onDeleteSpec(URI deletedSpecURI, String deletedSpecVersion, Properties headers) { if (topologySpecMap.containsKey(deletedSpecURI)) { topologySpecMap.remove(deletedSpecURI); } } @Override public synchronized void onUpdateSpec(Spec updatedSpec) { topologySpecMap.put(updatedSpec.getUri(), (TopologySpec) updatedSpec); } @Nonnull @Override public MetricContext getMetricContext() { return this.metricContext; } @Override public boolean isInstrumentationEnabled() { return null != this.metricContext; } @Override public List<Tag<?>> generateTags(State state){ return Collections.emptyList(); } @Override public void switchMetricContext(List<Tag<?>> tags) { throw new UnsupportedOperationException(); } @Override public void switchMetricContext(MetricContext context) { throw new UnsupportedOperationException(); } @Override public Map<URI, TopologySpec> getTopologySpecMap() { return this.topologySpecMap; } public abstract Dag<JobExecutionPlan> compileFlow(Spec spec); /** * Naive implementation of generating jobSpec, which fetch the first available template, * in an exemplified single-hop FlowCompiler implementation. * @param flowSpec * @return */ protected JobSpec jobSpecGenerator(FlowSpec flowSpec) { JobSpec jobSpec; JobSpec.Builder jobSpecBuilder = JobSpec.builder(jobSpecURIGenerator(flowSpec)) .withConfig(flowSpec.getConfig()) .withDescription(flowSpec.getDescription()) .withVersion(flowSpec.getVersion()); if (flowSpec.getTemplateURIs().isPresent() && templateCatalog.isPresent()) { // Only first template uri will be honored for Identity jobSpecBuilder = jobSpecBuilder.withTemplate(flowSpec.getTemplateURIs().get().iterator().next()); try { jobSpec = new ResolvedJobSpec(jobSpecBuilder.build(), templateCatalog.get()); log.info("Resolved JobSpec properties are: " + PropertiesUtils.prettyPrintProperties(jobSpec.getConfigAsProperties())); } catch (SpecNotFoundException | JobTemplate.TemplateException e) { throw new RuntimeException("Could not resolve template in JobSpec from TemplateCatalog", e); } } else { jobSpec = jobSpecBuilder.build(); log.info("Unresolved JobSpec properties are: " + jobSpec.getConfigAsProperties()); } // Remove schedule jobSpec.setConfig(jobSpec.getConfig().withoutPath(ConfigurationKeys.JOB_SCHEDULE_KEY)); // Add job.name and job.group if (flowSpec.getConfig().hasPath(ConfigurationKeys.FLOW_NAME_KEY)) { jobSpec.setConfig(jobSpec.getConfig() .withValue(ConfigurationKeys.JOB_NAME_KEY, flowSpec.getConfig().getValue(ConfigurationKeys.FLOW_NAME_KEY))); } if (flowSpec.getConfig().hasPath(ConfigurationKeys.FLOW_GROUP_KEY)) { jobSpec.setConfig(jobSpec.getConfig() .withValue(ConfigurationKeys.JOB_GROUP_KEY, flowSpec.getConfig().getValue(ConfigurationKeys.FLOW_GROUP_KEY))); } // Add flow execution id for this compilation long flowExecutionId = FlowUtils.getOrCreateFlowExecutionId(flowSpec); jobSpec.setConfig(jobSpec.getConfig().withValue(ConfigurationKeys.FLOW_EXECUTION_ID_KEY, ConfigValueFactory.fromAnyRef(flowExecutionId))); // Reset properties in Spec from Config jobSpec.setConfigAsProperties(ConfigUtils.configToProperties(jobSpec.getConfig())); return jobSpec; } /** * It can receive multiple number of parameters, needed to generate a unique URI. * Implementation is flowSpecCompiler dependent. * This method should return URI which has job name at third place, when split by "/" * e.g. /flowGroup/flowName * /flowGroup/flowName/sourceNode-targetNode * SafeDatasetCommit creates state store using this name and * {@link org.apache.gobblin.runtime.job_monitor.KafkaJobMonitor} extract job name to find the state store path. * @param objects * @return */ public URI jobSpecURIGenerator(Object... objects) { return ((FlowSpec)objects[0]).getUri(); } /** * It returns the template uri for job. * This method can be overridden by derived FlowToJobSpecCompiler classes. * @param flowSpec * @return template URI */ protected URI jobSpecTemplateURIGenerator(FlowSpec flowSpec) { // For now only first template uri will be honored for Identity return flowSpec.getTemplateURIs().get().iterator().next(); } }
3,921
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/flow/NoopDataMovementAuthorizer.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.flow; import com.typesafe.config.Config; import org.apache.gobblin.runtime.api.FlowSpec; import org.apache.gobblin.service.modules.flowgraph.DataNode; /** * {@link DataMovementAuthorizer} that always returns true. */ public class NoopDataMovementAuthorizer implements DataMovementAuthorizer { public NoopDataMovementAuthorizer(Config config) {} public boolean isMovementAuthorized(FlowSpec flowSpec, DataNode sourceNode, DataNode destNode) { return true; } }
3,922
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/flow/DataMovementAuthorizer.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.flow; import org.apache.gobblin.runtime.api.FlowSpec; import org.apache.gobblin.service.modules.flowgraph.DataNode; /** * Class that is called each time a flow is compiled to decide whether the data movement is authorized or not. */ public interface DataMovementAuthorizer { /** * Return true if the data movement is authorized given the flowspec and source/destination data node. */ public boolean isMovementAuthorized(FlowSpec flowSpec, DataNode sourceNode, DataNode destNode); }
3,923
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/flow/SpecCompiler.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.flow; import java.net.URI; import java.util.Map; import org.apache.gobblin.instrumented.Instrumentable; import org.apache.gobblin.runtime.api.Spec; import org.apache.gobblin.runtime.api.SpecCatalogListener; import org.apache.gobblin.runtime.api.SpecExecutor; import org.apache.gobblin.runtime.api.TopologySpec; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; /*** * Take in a logical {@link Spec} and compile corresponding materialized {@link Spec}s * and the mapping to {@link SpecExecutor} that they can be run on. */ public interface SpecCompiler extends SpecCatalogListener, Instrumentable { /*** * Take in a logical {@link Spec} and compile corresponding materialized {@link Spec}s * and the mapping to {@link SpecExecutor} that they can be run on. * All the specs generated from the compileFlow must have a * {@value org.apache.gobblin.configuration.ConfigurationKeys#FLOW_EXECUTION_ID_KEY} * @param spec {@link Spec} to compile. * @return Map of materialized physical {@link Spec} and {@link SpecExecutor}. */ Dag<JobExecutionPlan> compileFlow(Spec spec); /*** * Map of {@link Spec} URI and {@link TopologySpec} the {@link SpecCompiler} * is aware about. * @return Map of {@link Spec} URI and {@link TopologySpec} */ Map<URI, TopologySpec> getTopologySpecMap(); /** * Mark the {@link SpecCompiler} active/inactive. Useful to trigger the initialization of {@link SpecCompiler}, if * necessary, before it can start compiling {@link org.apache.gobblin.runtime.api.FlowSpec}s. * @param active */ void setActive(boolean active); /** * Waits for the {@link SpecCompiler} to become healthy. A {@link SpecCompiler} is healthy when all the component * services it depends on have been successfully initialized. For instance, the {@link MultiHopFlowCompiler} is healthy * when the {@link org.apache.gobblin.service.modules.flowgraph.DataNode}s and {@link org.apache.gobblin.service.modules.flowgraph.FlowEdge}s * can be added to the {@link org.apache.gobblin.service.modules.flowgraph.FlowGraph}. The {@link org.apache.gobblin.service.modules.flowgraph.FlowEdge} * instantiation in turn depends on the successful initialization of {@link org.apache.gobblin.runtime.spec_catalog.TopologyCatalog}, which * instantiates all the configured {@link SpecExecutor}s. */ public void awaitHealthy() throws InterruptedException; }
3,924
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/flow/FlowGraphPath.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.flow; import com.google.common.collect.Maps; import java.net.URI; import java.net.URISyntaxException; import java.util.ArrayList; import java.util.Iterator; import java.util.List; import java.util.Map; import java.util.Set; import java.util.stream.Collectors; import org.apache.hadoop.fs.Path; import com.google.common.annotations.VisibleForTesting; import com.google.common.base.Joiner; import com.google.common.base.Optional; import com.google.common.io.Files; import com.typesafe.config.Config; import com.typesafe.config.ConfigValueFactory; import lombok.Getter; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.runtime.api.FlowSpec; import org.apache.gobblin.runtime.api.JobSpec; import org.apache.gobblin.runtime.api.JobTemplate; import org.apache.gobblin.runtime.api.SpecExecutor; import org.apache.gobblin.runtime.api.SpecNotFoundException; import org.apache.gobblin.service.modules.dataset.DatasetDescriptor; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.flowgraph.Dag.DagNode; import org.apache.gobblin.service.modules.flowgraph.FlowEdge; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.service.modules.spec.JobExecutionPlanDagFactory; import org.apache.gobblin.service.modules.template.FlowTemplate; import org.apache.gobblin.util.ConfigUtils; /** * A class that encapsulates a path in the {@link org.apache.gobblin.service.modules.flowgraph.FlowGraph}. */ public class FlowGraphPath { @Getter private List<List<FlowEdgeContext>> paths; private final FlowSpec flowSpec; private final Long flowExecutionId; public FlowGraphPath(FlowSpec flowSpec, Long flowExecutionId) { this.flowSpec = flowSpec; this.flowExecutionId = flowExecutionId; } public void addPath(List<FlowEdgeContext> path) { if (this.paths == null) { this.paths = new ArrayList<>(); } this.paths.add(path); } /** * A method to convert a path of {@link FlowEdgeContext}s into a {@link Dag<JobExecutionPlan>}. * @param sysConfig containing environment config (e.g. metric/tracking event config) to be added to each {@link JobSpec}. * @return a {@link Dag<JobExecutionPlan>} * @throws SpecNotFoundException * @throws JobTemplate.TemplateException * @throws URISyntaxException */ public Dag<JobExecutionPlan> asDag(Config sysConfig) throws SpecNotFoundException, JobTemplate.TemplateException, URISyntaxException { Dag<JobExecutionPlan> flowDag = new Dag<>(new ArrayList<>()); for (List<FlowEdgeContext> path: paths) { Dag<JobExecutionPlan> pathDag = new Dag<>(new ArrayList<>()); Iterator<FlowEdgeContext> pathIterator = path.iterator(); while (pathIterator.hasNext()) { Dag<JobExecutionPlan> flowEdgeDag = convertHopToDag(pathIterator.next(), sysConfig); pathDag = concatenate(pathDag, flowEdgeDag); } flowDag = flowDag.merge(pathDag); } return flowDag; } /** * Concatenate two {@link Dag}s. Modify the {@link ConfigurationKeys#JOB_DEPENDENCIES} in the {@link JobSpec}s of the child * {@link Dag} to reflect the concatenation operation. * @param dagLeft The parent dag. * @param dagRight The child dag. * @return The concatenated dag with modified {@link ConfigurationKeys#JOB_DEPENDENCIES}. */ @VisibleForTesting static Dag<JobExecutionPlan> concatenate(Dag<JobExecutionPlan> dagLeft, Dag<JobExecutionPlan> dagRight) { //Compute the fork nodes - set of nodes with no dependents in the concatenated dag. Set<DagNode<JobExecutionPlan>> forkNodes = dagLeft.getEndNodes().stream(). filter(endNode -> isNodeForkable(endNode)).collect(Collectors.toSet()); Set<DagNode<JobExecutionPlan>> dependencyNodes = dagLeft.getDependencyNodes(forkNodes); if (!dependencyNodes.isEmpty()) { List<String> dependenciesList = dependencyNodes.stream() .map(dagNode -> dagNode.getValue().getJobSpec().getConfig().getString(ConfigurationKeys.JOB_NAME_KEY)) .collect(Collectors.toList()); String dependencies = Joiner.on(",").join(dependenciesList); for (DagNode<JobExecutionPlan> childNode : dagRight.getStartNodes()) { JobSpec jobSpec = childNode.getValue().getJobSpec(); jobSpec.setConfig(jobSpec.getConfig().withValue(ConfigurationKeys.JOB_DEPENDENCIES, ConfigValueFactory.fromAnyRef(dependencies))); } } return dagLeft.concatenate(dagRight, forkNodes); } private static boolean isNodeForkable(DagNode<JobExecutionPlan> dagNode) { Config jobConfig = dagNode.getValue().getJobSpec().getConfig(); return ConfigUtils.getBoolean(jobConfig, ConfigurationKeys.JOB_FORK_ON_CONCAT, false); } /** * Given an instance of {@link FlowEdge}, this method returns a {@link Dag < JobExecutionPlan >} that moves data * from the source of the {@link FlowEdge} to the destination of the {@link FlowEdge}. * @param flowEdgeContext an instance of {@link FlowEdgeContext}. * @param sysConfig environment config. * @return a {@link Dag} of {@link JobExecutionPlan}s associated with the {@link FlowEdge}. */ private Dag<JobExecutionPlan> convertHopToDag(FlowEdgeContext flowEdgeContext, Config sysConfig) throws SpecNotFoundException, JobTemplate.TemplateException, URISyntaxException { FlowTemplate flowTemplate = flowEdgeContext.getEdge().getFlowTemplate(); DatasetDescriptor inputDatasetDescriptor = flowEdgeContext.getInputDatasetDescriptor(); DatasetDescriptor outputDatasetDescriptor = flowEdgeContext.getOutputDatasetDescriptor(); Config mergedConfig = flowEdgeContext.getMergedConfig(); SpecExecutor specExecutor = flowEdgeContext.getSpecExecutor(); //Get resolved job configs from the flow template List<Config> resolvedJobConfigs = flowTemplate.getResolvedJobConfigs(mergedConfig, inputDatasetDescriptor, outputDatasetDescriptor); List<JobExecutionPlan> jobExecutionPlans = new ArrayList<>(resolvedJobConfigs.size()); Map<String, String> templateToJobNameMap = Maps.newHashMapWithExpectedSize(resolvedJobConfigs.size()); //Iterate over each resolved job config and convert the config to a JobSpec. for (Config resolvedJobConfig : resolvedJobConfigs) { JobExecutionPlan jobExecutionPlan = new JobExecutionPlan.Factory().createPlan(flowSpec, resolvedJobConfig, specExecutor, flowExecutionId, sysConfig); jobExecutionPlans.add(jobExecutionPlan); templateToJobNameMap.put(getJobTemplateName(jobExecutionPlan), jobExecutionPlan.getJobSpec().getConfig().getString( ConfigurationKeys.JOB_NAME_KEY)); } updateJobDependencies(jobExecutionPlans, templateToJobNameMap); return new JobExecutionPlanDagFactory().createDag(jobExecutionPlans); } /** * The job template name is derived from the {@link org.apache.gobblin.runtime.api.JobTemplate} URI. It is the * simple name of the path component of the URI. * @param jobExecutionPlan * @return the simple name of the job template from the URI of its path. */ private static String getJobTemplateName(JobExecutionPlan jobExecutionPlan) { Optional<URI> jobTemplateUri = jobExecutionPlan.getJobSpec().getTemplateURI(); if (jobTemplateUri.isPresent()) { return Files.getNameWithoutExtension(new Path(jobTemplateUri.get()).getName()); } else { return null; } } /** * A method to modify the {@link ConfigurationKeys#JOB_DEPENDENCIES} specified in a {@link JobTemplate} to those * which are usable in a {@link JobSpec}. * The {@link ConfigurationKeys#JOB_DEPENDENCIES} specified in a JobTemplate use the JobTemplate names * (i.e. the file names of the templates without the extension). However, the same {@link FlowTemplate} may be used * across multiple {@link FlowEdge}s. To ensure that we capture dependencies between jobs correctly as Dags from * successive hops are merged, we translate the {@link JobTemplate} name specified in the dependencies config to * {@link ConfigurationKeys#JOB_NAME_KEY} from the corresponding {@link JobSpec}, which is guaranteed to be globally unique. * For example, consider a {@link JobTemplate} with URI job1.job which has "job.dependencies=job2,job3" (where job2.job and job3.job are * URIs of other {@link JobTemplate}s). Also, let the job.name config for the three jobs (after {@link JobSpec} is compiled) be as follows: * "job.name=flowgrp1_flowName1_jobName1_1111", "job.name=flowgrp1_flowName1_jobName2_1121", and "job.name=flowgrp1_flowName1_jobName3_1131". Then, * for job1, this method will set "job.dependencies=flowgrp1_flowName1_jobName2_1121, flowgrp1_flowName1_jobName3_1131". * @param jobExecutionPlans a list of {@link JobExecutionPlan}s * @param templateToJobNameMap a HashMap that has the mapping from the {@link JobTemplate} names to job.name in corresponding * {@link JobSpec} */ private void updateJobDependencies(List<JobExecutionPlan> jobExecutionPlans, Map<String, String> templateToJobNameMap) { for (JobExecutionPlan jobExecutionPlan: jobExecutionPlans) { JobSpec jobSpec = jobExecutionPlan.getJobSpec(); if (jobSpec.getConfig().hasPath(ConfigurationKeys.JOB_DEPENDENCIES)) { List<String> jobDependencies = ConfigUtils.getStringList(jobSpec.getConfig(), ConfigurationKeys.JOB_DEPENDENCIES); List<String> updatedDependenciesList = new ArrayList<>(jobDependencies.size()); for (String dependency : jobDependencies) { if (!templateToJobNameMap.containsKey(dependency)) { //We should never hit this condition. The logic here is a safety check. throw new RuntimeException("TemplateToJobNameMap does not contain dependency " + dependency); } updatedDependenciesList.add(templateToJobNameMap.get(dependency)); } String updatedDependencies = Joiner.on(",").join(updatedDependenciesList); jobSpec.setConfig(jobSpec.getConfig().withValue(ConfigurationKeys.JOB_DEPENDENCIES, ConfigValueFactory.fromAnyRef(updatedDependencies))); } } } }
3,925
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/flow/MultiHopFlowCompiler.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.flow; import java.lang.reflect.InvocationTargetException; import java.net.URISyntaxException; import java.util.ArrayList; import java.util.HashMap; import java.util.List; import java.util.Map; import java.util.concurrent.CountDownLatch; import java.util.concurrent.TimeUnit; import java.util.concurrent.TimeoutException; import java.util.concurrent.atomic.AtomicReference; import java.util.concurrent.locks.ReadWriteLock; import java.util.concurrent.locks.ReentrantReadWriteLock; import java.util.stream.Collectors; import org.apache.commons.lang3.StringUtils; import org.apache.commons.lang3.reflect.ConstructorUtils; import org.apache.hadoop.fs.Path; import org.slf4j.Logger; import com.google.common.base.Joiner; import com.google.common.base.Optional; import com.google.common.base.Preconditions; import com.google.common.base.Splitter; import com.google.common.base.Throwables; import com.google.common.collect.Lists; import com.google.common.util.concurrent.ServiceManager; import com.typesafe.config.Config; import com.typesafe.config.ConfigValueFactory; import lombok.Getter; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.runtime.api.FlowSpec; import org.apache.gobblin.runtime.api.JobTemplate; import org.apache.gobblin.runtime.api.Spec; import org.apache.gobblin.runtime.api.SpecExecutor; import org.apache.gobblin.runtime.api.SpecNotFoundException; import org.apache.gobblin.service.ServiceConfigKeys; import org.apache.gobblin.service.modules.flowgraph.BaseFlowGraph; import org.apache.gobblin.service.modules.flowgraph.Dag; import org.apache.gobblin.service.modules.flowgraph.DataNode; import org.apache.gobblin.service.modules.flowgraph.DatasetDescriptorConfigKeys; import org.apache.gobblin.service.modules.flowgraph.FlowGraph; import org.apache.gobblin.service.modules.flowgraph.FlowGraphMonitor; import org.apache.gobblin.service.modules.flowgraph.pathfinder.PathFinder; import org.apache.gobblin.service.modules.restli.FlowConfigUtils; import org.apache.gobblin.service.modules.spec.JobExecutionPlan; import org.apache.gobblin.service.modules.template_catalog.ObservingFSFlowEdgeTemplateCatalog; import org.apache.gobblin.service.modules.template_catalog.UpdatableFSFlowTemplateCatalog; import org.apache.gobblin.service.monitoring.GitFlowGraphMonitor; import org.apache.gobblin.util.ClassAliasResolver; import org.apache.gobblin.util.ConfigUtils; /*** * Take in a logical {@link Spec} ie flow and compile corresponding materialized job {@link Spec} * and its mapping to {@link SpecExecutor}. */ @Alpha @Slf4j public class MultiHopFlowCompiler extends BaseFlowToJobSpecCompiler { private AtomicReference<FlowGraph> flowGraph; @Getter private ServiceManager serviceManager; @Getter private CountDownLatch initComplete = new CountDownLatch(1); private FlowGraphMonitor flowGraphMonitor; private ReadWriteLock rwLock = new ReentrantReadWriteLock(true); private DataMovementAuthorizer dataMovementAuthorizer; private Map<String, String> dataNodeAliasMap = new HashMap<>(); // a map to hold aliases of data nodes, e.g. gobblin.service.datanode.aliases.map=node1-dev:node1,node1-stg:node1,node1-prod:node1 public static final String DATA_NODE_ID_TO_ALIAS_MAP = ServiceConfigKeys.GOBBLIN_SERVICE_PREFIX + "datanode.aliases.map"; public MultiHopFlowCompiler(Config config) { this(config, true); } public MultiHopFlowCompiler(Config config, boolean instrumentationEnabled) { this(config, Optional.<Logger>absent(), instrumentationEnabled); } public MultiHopFlowCompiler(Config config, Optional<Logger> log) { this(config, log, true); } public MultiHopFlowCompiler(Config config, AtomicReference<FlowGraph> flowGraph) { super(config, Optional.absent(), true); this.flowGraph = flowGraph; this.dataMovementAuthorizer = new NoopDataMovementAuthorizer(config); } public MultiHopFlowCompiler(Config config, Optional<Logger> log, boolean instrumentationEnabled) { super(config, log, instrumentationEnabled); try { this.dataNodeAliasMap = config.hasPath(DATA_NODE_ID_TO_ALIAS_MAP) ? Splitter.on(",").withKeyValueSeparator(":").split(config.getString(DATA_NODE_ID_TO_ALIAS_MAP)) : new HashMap<>(); } catch (RuntimeException e) { MultiHopFlowCompiler.log.warn("Exception reading data node alias map, ignoring it.", e); } // Use atomic reference to avoid partial flowgraph upgrades during path compilation. this.flowGraph = new AtomicReference<>(new BaseFlowGraph(dataNodeAliasMap)); Optional<? extends UpdatableFSFlowTemplateCatalog> flowTemplateCatalog; if (config.hasPath(ServiceConfigKeys.TEMPLATE_CATALOGS_FULLY_QUALIFIED_PATH_KEY) && StringUtils.isNotBlank(config.getString(ServiceConfigKeys.TEMPLATE_CATALOGS_FULLY_QUALIFIED_PATH_KEY))) { try { String flowTemplateCatalogClassName = ConfigUtils.getString(this.config, ServiceConfigKeys.TEMPLATE_CATALOGS_CLASS_KEY, ObservingFSFlowEdgeTemplateCatalog.class.getCanonicalName()); flowTemplateCatalog = Optional.of( (UpdatableFSFlowTemplateCatalog) ConstructorUtils.invokeConstructor(Class.forName(new ClassAliasResolver<>(UpdatableFSFlowTemplateCatalog.class) .resolve(flowTemplateCatalogClassName)), config, rwLock)); } catch (NoSuchMethodException | IllegalAccessException | InvocationTargetException | InstantiationException | ClassNotFoundException e) { throw new RuntimeException("Cannot instantiate " + getClass().getName(), e); } } else { return; } Config gitFlowGraphConfig = this.config; if (this.config.hasPath(ConfigurationKeys.ENCRYPT_KEY_LOC)) { //Add encrypt.key.loc config to the config passed to GitFlowGraphMonitor gitFlowGraphConfig = this.config .withValue(GitFlowGraphMonitor.GIT_FLOWGRAPH_MONITOR_PREFIX + "." + ConfigurationKeys.ENCRYPT_KEY_LOC, config.getValue(ConfigurationKeys.ENCRYPT_KEY_LOC)); } try { String dataMovementAuthorizerClassName = ConfigUtils.getString(this.config, ServiceConfigKeys.DATA_MOVEMENT_AUTHORIZER_CLASS, NoopDataMovementAuthorizer.class.getCanonicalName()); this.dataMovementAuthorizer = (DataMovementAuthorizer) ConstructorUtils.invokeConstructor(Class.forName(new ClassAliasResolver<>(DataMovementAuthorizer.class).resolve(dataMovementAuthorizerClassName)), this.config); } catch (NoSuchMethodException | IllegalAccessException | InvocationTargetException | InstantiationException | ClassNotFoundException e) { throw new RuntimeException(e); } try { String flowGraphMonitorClassName = ConfigUtils.getString(this.config, ServiceConfigKeys.GOBBLIN_SERVICE_FLOWGRAPH_CLASS_KEY, GitFlowGraphMonitor.class.getCanonicalName()); this.flowGraphMonitor = (FlowGraphMonitor) ConstructorUtils.invokeConstructor(Class.forName(new ClassAliasResolver<>(FlowGraphMonitor.class).resolve( flowGraphMonitorClassName)), gitFlowGraphConfig, flowTemplateCatalog, this, this.topologySpecMap, this.getInitComplete(), instrumentationEnabled); } catch (NoSuchMethodException | IllegalAccessException | InvocationTargetException | InstantiationException | ClassNotFoundException e) { throw new RuntimeException(e); } this.serviceManager = (flowTemplateCatalog.isPresent() && flowTemplateCatalog.get() instanceof ObservingFSFlowEdgeTemplateCatalog) ? new ServiceManager(Lists.newArrayList(this.flowGraphMonitor, flowTemplateCatalog.get())) : new ServiceManager(Lists.newArrayList(this.flowGraphMonitor)); addShutdownHook(); //Start the git flow graph monitor try { this.serviceManager.startAsync().awaitHealthy(5, TimeUnit.SECONDS); } catch (TimeoutException te) { MultiHopFlowCompiler.log.error("Timed out while waiting for the service manager to start up", te); throw new RuntimeException(te); } } /** * Mark the {@link SpecCompiler} as active. This in turn activates the {@link GitFlowGraphMonitor}, allowing to start polling * and processing changes * @param active */ @Override public void setActive(boolean active) { super.setActive(active); if (this.flowGraphMonitor != null) { this.flowGraphMonitor.setActive(active); } } @Override public void awaitHealthy() throws InterruptedException { if (this.getInitComplete().getCount() > 0) { log.info("Waiting for the MultiHopFlowCompiler to become healthy.."); this.getInitComplete().await(); log.info("The MultihopFlowCompiler is healthy and ready to orchestrate flows."); } return; } /** * j * @param spec an instance of {@link FlowSpec}. * @return A DAG of {@link JobExecutionPlan}s, which encapsulates the compiled {@link org.apache.gobblin.runtime.api.JobSpec}s * together with the {@link SpecExecutor} where the job can be executed. */ @Override public Dag<JobExecutionPlan> compileFlow(Spec spec) { Preconditions.checkNotNull(spec); Preconditions.checkArgument(spec instanceof FlowSpec, "MultiHopFlowCompiler only accepts FlowSpecs"); FlowGraph graph = this.flowGraph.get(); long startTime = System.nanoTime(); FlowSpec flowSpec = (FlowSpec) spec; String source = FlowConfigUtils.getDataNode(flowSpec.getConfig(), ServiceConfigKeys.FLOW_SOURCE_IDENTIFIER_KEY, this.dataNodeAliasMap); String destination = FlowConfigUtils.getDataNode(flowSpec.getConfig(), ServiceConfigKeys.FLOW_DESTINATION_IDENTIFIER_KEY, this.dataNodeAliasMap); DataNode sourceNode = graph.getNode(source); if (sourceNode == null) { flowSpec.addCompilationError(source, destination, String.format("Flowgraph does not have a node with id %s", source)); return null; } List<String> destNodeIds = FlowConfigUtils.getDataNodes(flowSpec.getConfig(), ServiceConfigKeys.FLOW_DESTINATION_IDENTIFIER_KEY, this.dataNodeAliasMap); List<DataNode> destNodes = destNodeIds.stream().map(graph::getNode).collect(Collectors.toList()); if (destNodes.contains(null)) { flowSpec.addCompilationError(source, destination, String.format("Flowgraph does not have a node with id %s", destNodeIds.get(destNodes.indexOf(null)))); return null; } log.info(String.format("Compiling flow for source: %s and destination: %s", source, destination)); List<FlowSpec> flowSpecs = splitFlowSpec(flowSpec); Dag<JobExecutionPlan> jobExecutionPlanDag = new Dag<>(new ArrayList<>()); try { this.rwLock.readLock().lock(); for (FlowSpec datasetFlowSpec : flowSpecs) { for (DataNode destNode : destNodes) { long authStartTime = System.nanoTime(); try { boolean authorized = this.dataMovementAuthorizer.isMovementAuthorized(flowSpec, sourceNode, destNode); Instrumented.updateTimer(dataAuthorizationTimer, System.nanoTime() - authStartTime, TimeUnit.NANOSECONDS); if (!authorized) { String message = String.format("Data movement is not authorized for flow: %s, source: %s, destination: %s", flowSpec.getUri().toString(), source, destination); log.error(message); datasetFlowSpec.addCompilationError(source, destination, message); return null; } } catch (Exception e) { Instrumented.markMeter(flowCompilationFailedMeter); datasetFlowSpec.addCompilationError(source, destination, Throwables.getStackTraceAsString(e)); return null; } } //Compute the path from source to destination. FlowGraphPath flowGraphPath = graph.findPath(datasetFlowSpec); if (flowGraphPath != null) { //Convert the path into a Dag of JobExecutionPlans. jobExecutionPlanDag = jobExecutionPlanDag.merge(flowGraphPath.asDag(this.config)); } } if (jobExecutionPlanDag.isEmpty()) { Instrumented.markMeter(flowCompilationFailedMeter); String message = String.format("No path found from source: %s and destination: %s", source, destination); log.info(message); if (!flowSpec.getCompilationErrors().stream().anyMatch(compilationError -> compilationError.errorPriority == 0)) { flowSpec.addCompilationError(source, destination, message); } return null; } } catch (PathFinder.PathFinderException | SpecNotFoundException | JobTemplate.TemplateException | URISyntaxException | ReflectiveOperationException e) { Instrumented.markMeter(flowCompilationFailedMeter); String message = String.format("Exception encountered while compiling flow for source: %s and destination: %s, %s", source, destination, Throwables.getStackTraceAsString(e)); log.error(message, e); flowSpec.addCompilationError(source, destination, message); return null; } finally { this.rwLock.readLock().unlock(); } Instrumented.markMeter(flowCompilationSuccessFulMeter); Instrumented.updateTimer(flowCompilationTimer, System.nanoTime() - startTime, TimeUnit.NANOSECONDS); // Clear compilation errors now that compilation is successful flowSpec.clearCompilationErrors(); return jobExecutionPlanDag; } public void setFlowGraph(FlowGraph flowGraph) { this.flowGraph.set(flowGraph); } /** * If {@link FlowSpec} has {@link ConfigurationKeys#DATASET_SUBPATHS_KEY}, split it into multiple flowSpecs using a * provided base input and base output path to generate multiple source/destination paths. */ private static List<FlowSpec> splitFlowSpec(FlowSpec flowSpec) { long flowExecutionId = FlowUtils.getOrCreateFlowExecutionId(flowSpec); List<FlowSpec> flowSpecs = new ArrayList<>(); Config flowConfig = flowSpec.getConfig(); if (flowConfig.hasPath(ConfigurationKeys.DATASET_SUBPATHS_KEY)) { List<String> datasetSubpaths = ConfigUtils.getStringList(flowConfig, ConfigurationKeys.DATASET_SUBPATHS_KEY); String baseInputPath = ConfigUtils.getString(flowConfig, ConfigurationKeys.DATASET_BASE_INPUT_PATH_KEY, "/"); String baseOutputPath = ConfigUtils.getString(flowConfig, ConfigurationKeys.DATASET_BASE_OUTPUT_PATH_KEY, "/"); if (ConfigUtils.getBoolean(flowConfig, ConfigurationKeys.DATASET_COMBINE_KEY, false)) { Config newConfig = flowConfig.withoutPath(ConfigurationKeys.DATASET_SUBPATHS_KEY) .withValue(DatasetDescriptorConfigKeys.FLOW_INPUT_DATASET_DESCRIPTOR_PREFIX + "." + DatasetDescriptorConfigKeys.PATH_KEY, ConfigValueFactory.fromAnyRef(baseInputPath)) .withValue(DatasetDescriptorConfigKeys.FLOW_OUTPUT_DATASET_DESCRIPTOR_PREFIX + "." + DatasetDescriptorConfigKeys.PATH_KEY, ConfigValueFactory.fromAnyRef(baseOutputPath)) .withValue(DatasetDescriptorConfigKeys.FLOW_INPUT_DATASET_DESCRIPTOR_PREFIX + ".subPaths", ConfigValueFactory.fromAnyRef(convertStringListToGlobPattern(datasetSubpaths))) .withValue(DatasetDescriptorConfigKeys.FLOW_OUTPUT_DATASET_DESCRIPTOR_PREFIX + ".subPaths", ConfigValueFactory.fromAnyRef(convertStringListToGlobPattern(datasetSubpaths))); flowSpecs.add(copyFlowSpecWithNewConfig(flowSpec, newConfig)); } else { for (String subPath : datasetSubpaths) { Config newConfig = flowConfig.withoutPath(ConfigurationKeys.DATASET_SUBPATHS_KEY) .withValue(ConfigurationKeys.FLOW_EXECUTION_ID_KEY, ConfigValueFactory.fromAnyRef(flowExecutionId)) .withValue(DatasetDescriptorConfigKeys.FLOW_INPUT_DATASET_DESCRIPTOR_PREFIX + "." + DatasetDescriptorConfigKeys.PATH_KEY, ConfigValueFactory.fromAnyRef(new Path(baseInputPath, subPath).toString())) .withValue(DatasetDescriptorConfigKeys.FLOW_OUTPUT_DATASET_DESCRIPTOR_PREFIX + "." + DatasetDescriptorConfigKeys.PATH_KEY, ConfigValueFactory.fromAnyRef(new Path(baseOutputPath, subPath).toString())); flowSpecs.add(copyFlowSpecWithNewConfig(flowSpec, newConfig)); } } } else { flowSpecs.add(flowSpec); } return flowSpecs; } /** * Convert string list to string pattern that will work for globs. * * e.g. ["test1", "test2", test3"] -> "{test1,test2,test}" */ private static String convertStringListToGlobPattern(List<String> stringList) { return "{" + Joiner.on(",").join(stringList) + "}"; } private static FlowSpec copyFlowSpecWithNewConfig(FlowSpec flowSpec, Config newConfig) { FlowSpec.Builder builder = FlowSpec.builder(flowSpec.getUri()).withVersion(flowSpec.getVersion()) .withDescription(flowSpec.getDescription()).withConfig(newConfig); if (flowSpec.getTemplateURIs().isPresent()) { builder = builder.withTemplates(flowSpec.getTemplateURIs().get()); } if (flowSpec.getChildSpecs().isPresent()) { builder = builder.withTemplates(flowSpec.getChildSpecs().get()); } return builder.build(); } /** * Register a shutdown hook for this thread. */ private void addShutdownHook() { ServiceManager manager = this.serviceManager; Runtime.getRuntime().addShutdownHook(new Thread() { public void run() { // Give the services 5 seconds to stop to ensure that we are responsive to shutdown // requests. try { manager.stopAsync().awaitStopped(5, TimeUnit.SECONDS); } catch (TimeoutException timeout) { // stopping timed out } } }); } }
3,926
0
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules
Create_ds/gobblin/gobblin-service/src/main/java/org/apache/gobblin/service/modules/flow/FlowUtils.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.service.modules.flow; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.runtime.api.FlowSpec; public class FlowUtils { /** * A FlowSpec contains a FlowExecutionId if it is a runOnce flow. * Refer {@link FlowConfigResourceLocalHandler#createFlowSpecForConfig} for details. * @param spec flow spec * @return flow execution id */ public static long getOrCreateFlowExecutionId(FlowSpec spec) { long flowExecutionId; if (spec.getConfig().hasPath(ConfigurationKeys.FLOW_EXECUTION_ID_KEY)) { flowExecutionId = spec.getConfig().getLong(ConfigurationKeys.FLOW_EXECUTION_ID_KEY); } else { flowExecutionId = System.currentTimeMillis(); } return flowExecutionId; } }
3,927
0
Create_ds/gobblin/gobblin-binary-management/src/test/java/org/apache/gobblin
Create_ds/gobblin/gobblin-binary-management/src/test/java/org/apache/gobblin/binary_creation/OrcTestToolsTest.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.binary_creation; import com.google.common.io.Files; import java.io.File; import org.apache.avro.Schema; import org.apache.avro.SchemaBuilder; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo; import org.apache.orc.TypeDescription; import org.testng.Assert; import org.testng.annotations.Test; public class OrcTestToolsTest { public DataTestTools orcTools = new OrcTestTools();; @Test public void test() throws Exception { String resourceName = "orcWriterTest"; File tmpDir = Files.createTempDir(); FileSystem fs = FileSystem.get(new Configuration()); Path output = new Path(tmpDir.getAbsolutePath(), "test"); orcTools.writeJsonResourceRecordsAsBinary(resourceName, null, output, null); Assert.assertTrue(orcTools.checkSameFilesAndRecords(orcTools.readAllRecordsInJsonResource(resourceName, null), orcTools.readAllRecordsInBinaryDirectory(fs, output), true, null, false)); } @Test public void testSchemaToTypeInfoConversion() throws Exception { // Simple non-nested case: Schema avroSchema = SchemaBuilder.record("test") .fields() .name("id") .type() .intType() .noDefault() .name("timestamp") .type() .stringType() .noDefault() .endRecord(); TypeInfo orcSchema = OrcTestTools.convertAvroSchemaToOrcSchema(avroSchema); String targetOrcSchemaString = "struct<id:int,timestamp:string>"; Assert.assertEquals(targetOrcSchemaString, orcSchema.toString()); // Nested case: avroSchema = SchemaBuilder.record("nested") .fields() .name("nestedId") .type() .array() .items() .stringType() .noDefault() .name("timestamp") .type() .stringType() .noDefault() .endRecord(); orcSchema = OrcTestTools.convertAvroSchemaToOrcSchema(avroSchema); TypeDescription targetTypeDescription = TypeDescription.createStruct() .addField("nestedId", TypeDescription.createList(TypeDescription.createString())) .addField("timestamp", TypeDescription.createString()); Assert.assertEquals(orcSchema.toString().toLowerCase(), targetTypeDescription.toString().toLowerCase()); } }
3,928
0
Create_ds/gobblin/gobblin-binary-management/src/test/java/org/apache/gobblin
Create_ds/gobblin/gobblin-binary-management/src/test/java/org/apache/gobblin/binary_creation/AvroTestToolsTest.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.binary_creation; import com.google.common.io.Files; import java.io.File; import org.apache.avro.Schema; import org.apache.avro.generic.GenericData; import org.apache.avro.generic.GenericRecordBuilder; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.testng.Assert; import org.testng.annotations.Test; import org.apache.gobblin.binary_creation.AvroTestTools.*; public class AvroTestToolsTest { @Test public void test() throws Exception { DataTestTools testTools = new AvroTestTools(); String resourceName = "avroWriterTest"; File tmpDir = Files.createTempDir(); FileSystem fs = FileSystem.getLocal(new Configuration()); Path output = new Path(tmpDir.getAbsolutePath(), "test"); testTools.writeJsonResourceRecordsAsBinary(resourceName, fs, output, null); Assert.assertTrue(testTools.checkSameFilesAndRecords(testTools.readAllRecordsInJsonResource(resourceName, null), testTools.readAllRecordsInBinaryDirectory(fs, output), false, null, true)); } @Test public void testGenericRecordDataComparisonWithoutSchema() throws Exception { Schema avroSchema = (new Schema.Parser()).parse( "{\n" + " \"namespace\": \"com.linkedin.compliance.test\",\n" + " \"type\": \"record\",\n" + " \"name\": \"SimpleTest\",\n" + " \"fields\": [\n" + " {\n" + " \"name\": \"memberId\",\n" + " \"type\": \"int\"\n" + " },\n" + " {\n" + " \"name\": \"name\",\n" + " \"type\": \"string\"\n" + " }\n" + " ]\n" + "}"); Schema avroSchemaDiffInNamespace = (new Schema.Parser()).parse( "{\n" + " \"namespace\": \"com.linkedin.whatever\",\n" + " \"type\": \"record\",\n" + " \"name\": \"SimpleTest\",\n" + " \"fields\": [\n" + " {\n" + " \"name\": \"memberId\",\n" + " \"type\": \"int\"\n" + " },\n" + " {\n" + " \"name\": \"name\",\n" + " \"type\": \"string\"\n" + " }\n" + " ]\n" + "}"); Schema nullableSchema = (new Schema.Parser()).parse( "{\n" + " \"namespace\": \"com.linkedin.compliance.test\",\n" + " \"type\": \"record\",\n" + " \"name\": \"SimpleTest\",\n" + " \"fields\": [\n" + " {\n" + " \"name\": \"memberId\",\n" + " \"type\": [\n" + " \"null\",\n" + " \"int\",\n" + " \"string\"\n" + " ]\n" + " },\n" + " {\n" + " \"name\": \"name\",\n" + " \"type\": \"string\"\n" + " }\n" + " ]\n" + "}"); GenericRecordBuilder builder_0 = new GenericRecordBuilder(avroSchema); builder_0.set("memberId", "1"); builder_0.set("name", "alice"); GenericData.Record record_0 = builder_0.build(); GenericRecordBuilder builder_1 = new GenericRecordBuilder(avroSchemaDiffInNamespace); builder_1.set("memberId", "1"); builder_1.set("name", "alice"); GenericData.Record record_1 = builder_1.build(); GenericRecordBuilder builder_2 = new GenericRecordBuilder(avroSchemaDiffInNamespace); builder_2.set("memberId", "1"); builder_2.set("name", "alice"); GenericData.Record record_2 = builder_2.build(); GenericRecordBuilder builder_3 = new GenericRecordBuilder(avroSchemaDiffInNamespace); builder_3.set("memberId", "2"); builder_3.set("name", "bob"); GenericData.Record record_3 = builder_3.build(); GenericRecordBuilder builder_4 = new GenericRecordBuilder(nullableSchema); builder_4.set("memberId", null); builder_4.set("name", "bob"); GenericData.Record record_4 = builder_4.build(); GenericRecordBuilder builder_5 = new GenericRecordBuilder(nullableSchema); builder_5.set("memberId", null); builder_5.set("name", "bob"); GenericData.Record record_5 = builder_5.build(); Assert.assertTrue(!record_0.equals(record_1)); AvroTestTools.GenericRecordWrapper wrapper_0 = new GenericRecordWrapper(record_0); GenericRecordWrapper wrapper_1 = new GenericRecordWrapper(record_1); GenericRecordWrapper wrapper_2 = new GenericRecordWrapper(record_2); GenericRecordWrapper wrapper_3 = new GenericRecordWrapper(record_3); GenericRecordWrapper wrapper_4 = new GenericRecordWrapper(record_4); GenericRecordWrapper wrapper_5 = new GenericRecordWrapper(record_5); Assert.assertEquals(wrapper_0, wrapper_1); Assert.assertEquals(wrapper_1, wrapper_2); Assert.assertNotSame(wrapper_2, wrapper_3); Assert.assertEquals(wrapper_4, wrapper_5); } }
3,929
0
Create_ds/gobblin/gobblin-binary-management/src/main/java/org/apache/gobblin
Create_ds/gobblin/gobblin-binary-management/src/main/java/org/apache/gobblin/binary_creation/AvroTestTools.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.binary_creation; import com.google.common.collect.AbstractIterator; import java.io.File; import java.io.IOException; import java.io.InputStream; import java.util.Collection; import java.util.Iterator; import java.util.List; import java.util.Map; import java.util.Objects; import java.util.Set; import java.util.Spliterators; import java.util.TreeMap; import java.util.function.Function; import java.util.stream.Collectors; import java.util.stream.StreamSupport; import javax.annotation.Nullable; import lombok.AllArgsConstructor; import lombok.Getter; import lombok.experimental.Delegate; import lombok.extern.slf4j.Slf4j; import org.apache.avro.Schema; import org.apache.avro.file.DataFileReader; import org.apache.avro.file.DataFileWriter; import org.apache.avro.file.SeekableInput; import org.apache.avro.generic.GenericDatumReader; import org.apache.avro.generic.GenericDatumWriter; import org.apache.avro.generic.GenericRecord; import org.apache.avro.io.Decoder; import org.apache.avro.io.DecoderFactory; import org.apache.avro.mapred.FsInput; import org.apache.gobblin.util.FileListUtils; import org.apache.gobblin.util.PathUtils; import org.apache.gobblin.util.filters.HiddenFilter; import org.apache.hadoop.fs.FileStatus; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.apache.hadoop.fs.PathFilter; import org.reflections.Reflections; import org.reflections.scanners.ResourcesScanner; import org.reflections.util.ConfigurationBuilder; /** * A implementation of {@link DataTestTools} for Avro format. */ @Slf4j public class AvroTestTools extends DataTestTools<AvroTestTools.RecordIterator, Schema> { public boolean checkSameFilesAndRecords(TreeMap<String, RecordIterator> expected, TreeMap<String, RecordIterator> observed, boolean allowDifferentOrder, Collection<String> blacklistRecordFields, boolean allowDifferentSchema) { Iterator<String> keys1 = expected.navigableKeySet().iterator(); Iterator<String> keys2 = observed.navigableKeySet().iterator(); return compareIterators(keys1, keys2, (key1, key2) -> { if (!removeExtension(key1).equals(removeExtension(key2))) { log.error(String.format("Mismatched files: %s and %s", key1, key2)); return false; } RecordIterator it1 = expected.get(key1); RecordIterator it2 = observed.get(key2); if (!allowDifferentSchema && !it1.getSchema().equals(it2.getSchema())) { log.error(String.format("Mismatched schemas: %s and %s", key1, key2)); return false; } if (allowDifferentOrder) { Set r1 = allowDifferentSchema ? toSetWithBlacklistedFields(it1, blacklistRecordFields, GenericRecordWrapper::new) : toSetWithBlacklistedFields(it1, blacklistRecordFields, Function.identity()); Set r2 = allowDifferentSchema ? toSetWithBlacklistedFields(it2, blacklistRecordFields, GenericRecordWrapper::new) : toSetWithBlacklistedFields(it2, blacklistRecordFields, Function.identity()); if (r1.equals(r2)) { return true; } else { log.info("Sets of records differ."); return false; } } else { return compareIterators(it1, it2, (r1, r2) -> { if (blacklistRecordFields != null) { for (String blacklisted : blacklistRecordFields) { r1.put(blacklisted, null); r2.put(blacklisted, null); } } return allowDifferentSchema ? GenericRecordWrapper.compareGenericRecordRegardlessOfSchema(r1, r2) : r1.equals(r2); }); } }); } private static <T> Set<T> toSetWithBlacklistedFields(Iterator<GenericRecord> it, Collection<String> blacklistRecordFields, Function<GenericRecord, T> transform) { return StreamSupport.stream(Spliterators.spliteratorUnknownSize(it, 0), false).map(r -> { for (String blacklisted : blacklistRecordFields) { r.put(blacklisted, null); } return transform.apply(r); }).collect(Collectors.toSet()); } /** * Read all avro records in an HDFS location into a map from file name to {@link RecordIterator}. */ @Override public TreeMap<String, RecordIterator> readAllRecordsInBinaryDirectory(FileSystem fs, Path path) throws IOException { TreeMap<String, RecordIterator> output = new TreeMap<>(); if (!fs.exists(path)) { return output; } PathFilter pathFilter = new HiddenFilter(); for (FileStatus status : FileListUtils.listFilesRecursively(fs, path, pathFilter)) { SeekableInput sin = new FsInput(status.getPath(), fs); DataFileReader<GenericRecord> dfr = new DataFileReader<>(sin, new GenericDatumReader<>()); String key = PathUtils.relativizePath(status.getPath(), path).toString(); output.put(key, new RecordIterator(dfr.getSchema(), new AbstractIterator<GenericRecord>() { @Override protected GenericRecord computeNext() { if (dfr.hasNext()) { return dfr.next(); } else { try { dfr.close(); } catch (IOException ioe) { log.error("Failed to close data file reader.", ioe); } endOfData(); return null; } } })); } return output; } /** * Read all avro records in a json base resource in classpath into a map from file name to {@link RecordIterator}. */ @Override public TreeMap<String, RecordIterator> readAllRecordsInJsonResource(String baseResource, @Nullable Schema schema) throws IOException { if (schema == null) { String schemaResource = new File(baseResource, "schema.avsc").toString(); schema = readAvscSchema(schemaResource, AvroTestTools.class); } TreeMap<String, RecordIterator> output = new TreeMap<>(); for (String file : getJsonFileSetByResourceRootName(baseResource)) { log.info("Reading json record from " + file); String name = PathUtils.relativizePath(new Path(file), new Path(baseResource)).toString(); String schemaResourceName = new File(new File(file).getParent(), "schema.avsc").toString(); Schema thisSchema = readAvscSchema(schemaResourceName, AvroTestTools.class); Schema actualSchema = thisSchema == null ? schema : thisSchema; try (InputStream is = AvroTestTools.class.getClassLoader().getResourceAsStream(file)) { output.put(name, readRecordsFromJsonInputStream(actualSchema, is, DecoderFactory.get().jsonDecoder(actualSchema, is))); } } return output; } private static RecordIterator readRecordsFromJsonInputStream(Schema schema, InputStream is, Decoder decoder) { GenericDatumReader<GenericRecord> reader = new GenericDatumReader<>(schema); return new RecordIterator(schema, new AbstractIterator<GenericRecord>() { @Override protected GenericRecord computeNext() { try { return reader.read(null, decoder); } catch (IOException ioe) { try { is.close(); } catch (IOException exc) { log.warn("Failed to close input stream.", exc); } endOfData(); return null; } } }); } /** * Materialize records in a classpath package into HDFS avro records. * @param baseResource name of the package. The package should contain the following: * - Exactly one resource called <name>.avsc containing the schema of the records * (or an explicit schema passed as an argument). * - One or more data files called *.json containing the records. * @param fs the {@link FileSystem} where the records will be written. * @param targetPath the path where the records will be written. * @param schema Schema of the records, or null to read automatically from a resource. * @throws IOException */ public Schema writeJsonResourceRecordsAsBinary(String baseResource, FileSystem fs, Path targetPath, @Nullable Schema schema) throws IOException { TreeMap<String, RecordIterator> recordMap = readAllRecordsInJsonResource(baseResource, schema); Schema outputSchema = recordMap.lastEntry().getValue().getSchema(); for (Map.Entry<String, RecordIterator> entry : recordMap.entrySet()) { writeAsAvroBinary(entry.getValue(), entry.getValue().getSchema(), fs, new Path(targetPath, removeExtension(entry.getKey()) + ".avro")); } return outputSchema; } /** * Read schema from an avsc resource file. */ public static Schema readAvscSchema(String resource, Class loadedClass) throws IOException { try (InputStream is = loadedClass.getClassLoader().getResourceAsStream(resource)) { return is != null ? new Schema.Parser().parse(is) : null; } } private void writeAsAvroBinary(Iterator<GenericRecord> input, Schema schema, FileSystem fs, Path outputPath) throws IOException { DataFileWriter writer = new DataFileWriter(new GenericDatumWriter()); writer.create(schema, fs.create(outputPath, true)); while (input.hasNext()) { writer.append(input.next()); } writer.close(); log.info("Successfully wrote avro file to path " + outputPath); } /** * An iterator over {@link GenericRecord} which is also aware of schema. */ @AllArgsConstructor public static class RecordIterator implements Iterator<GenericRecord> { @Getter private final Schema schema; @Delegate private final Iterator<GenericRecord> it; } /** * A wrapper of {@link GenericRecord} when schema of record is not important in comparison. */ @AllArgsConstructor public static class GenericRecordWrapper { public GenericRecord record; @Override public boolean equals(Object o) { if (this == o) { return true; } if (o == null || getClass() != o.getClass()) { return false; } GenericRecordWrapper that = (GenericRecordWrapper) o; return compareGenericRecordRegardlessOfSchema(record, that.record); } @Override public int hashCode() { // Getting value object array int indexLen = record.getSchema().getFields().size(); Object[] objArr = new Object[indexLen]; for (int i = 0; i < indexLen; i++) { objArr[i] = record.get(i); } return Objects.hash(objArr); } /** * Compare two {@link GenericRecord} instance without considering their schema. * Useful when we want to compare two records by discarding some of fields like header. */ static boolean compareGenericRecordRegardlessOfSchema(GenericRecord r1, GenericRecord r2) { List<Schema.Field> listOfFields1 = r1.getSchema().getFields(); List<Schema.Field> listOfFields2 = r2.getSchema().getFields(); if (listOfFields1.size() != listOfFields2.size()) { return false; } boolean result = true; for (int i = 0; i < listOfFields1.size(); i++) { result = result && ( ((r1.get(i) == null && r2.get(i) == null) || (listOfFields1.get(i).name().equals(listOfFields2.get(i).name()) && (r1.get(i).equals(r2.get(i))))) ); } return result; } } // Package-private methods shared by different format's tool-kit. static String removeExtension(String string) { if (string.endsWith(".avro") || string.endsWith(".json")) { return string.substring(0, string.length() - 5); } throw new IllegalArgumentException("Only support avro and json extensions."); } static Set<String> getJsonFileSetByResourceRootName(String baseResource) { Reflections reflections = new Reflections(new ConfigurationBuilder() .forPackages(baseResource) .filterInputsBy(name -> name.startsWith(baseResource)) .setScanners(new ResourcesScanner())); return reflections.getResources(url -> url.endsWith(".json")); } public static boolean isResourceExisted(String resource) throws IOException { return AvroTestTools.class.getClassLoader().getResource(resource) != null; } }
3,930
0
Create_ds/gobblin/gobblin-binary-management/src/main/java/org/apache/gobblin
Create_ds/gobblin/gobblin-binary-management/src/main/java/org/apache/gobblin/binary_creation/OrcTestTools.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.binary_creation; import com.google.common.collect.AbstractIterator; import java.io.DataInput; import java.io.DataOutput; import java.io.File; import java.io.IOException; import java.io.InputStream; import java.lang.reflect.Field; import java.util.ArrayList; import java.util.Collection; import java.util.Iterator; import java.util.Map; import java.util.Properties; import java.util.TreeMap; import javax.annotation.Nullable; import lombok.AllArgsConstructor; import lombok.Getter; import lombok.experimental.Delegate; import lombok.extern.slf4j.Slf4j; import org.apache.avro.Schema; import org.apache.avro.generic.GenericDatumReader; import org.apache.avro.generic.GenericRecord; import org.apache.avro.io.Decoder; import org.apache.avro.io.DecoderFactory; import org.apache.gobblin.util.FileListUtils; import org.apache.gobblin.util.PathUtils; import org.apache.gobblin.util.filters.HiddenFilter; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.FileStatus; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.apache.hadoop.fs.PathFilter; import org.apache.hadoop.hive.ql.io.orc.OrcFile; import org.apache.hadoop.hive.ql.io.orc.OrcStruct; import org.apache.hadoop.hive.ql.io.orc.Reader; import org.apache.hadoop.hive.ql.io.orc.RecordReader; import org.apache.hadoop.hive.ql.io.orc.Writer; import org.apache.hadoop.hive.serde2.SerDeException; import org.apache.hadoop.hive.serde2.avro.AvroGenericRecordWritable; import org.apache.hadoop.hive.serde2.avro.AvroObjectInspectorGenerator; import org.apache.hadoop.hive.serde2.avro.AvroSerDe; import org.apache.hadoop.hive.serde2.avro.AvroSerdeUtils; import org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector; import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo; import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoUtils; import org.apache.hadoop.io.DoubleWritable; import org.apache.hadoop.io.IntWritable; import org.apache.hadoop.io.LongWritable; import org.apache.hadoop.io.ShortWritable; import org.apache.hadoop.io.Text; import org.apache.hadoop.io.Writable; import static org.apache.gobblin.binary_creation.AvroTestTools.*; // A class that examines ORC-Format file in Purger Integration test. @Slf4j public class OrcTestTools extends DataTestTools<OrcTestTools.OrcRowIterator, TypeInfo> { /** * * @param expected * @param observed * @param allowDifferentOrder ORC tools will not use this parameter currently. * @param blacklistRecordFields ORC tools will not use this parameter currently. * @return If two sets of files are identical. * Note that there might be an ordering issue in this comparison method. When one is drafting an ORC integration * test, try to name all json files differently. */ @Override public boolean checkSameFilesAndRecords(TreeMap<String, OrcRowIterator> expected, TreeMap<String, OrcRowIterator> observed, boolean allowDifferentOrder, Collection<String> blacklistRecordFields, boolean allowDifferentSchema) { Iterator<String> keys1 = expected.navigableKeySet().iterator(); Iterator<String> keys2 = observed.navigableKeySet().iterator(); return compareIterators(keys1, keys2, (key1, key2) -> { // ORC file doesn't have extension by Linkedin's convention. if (!removeExtension(key1).equals(key2)) { log.error(String.format("Mismatched files: %s and %s", key1, key2)); return false; } OrcRowIterator it1 = expected.get(key1); OrcRowIterator it2 = observed.get(key2); if (!it1.getTypeInfo().equals(it2.getTypeInfo())) { log.error(String.format("Mismatched Typeinfo: %s and %s", key1, key2)); return false; } boolean result = true; while (it1.hasNext()) { if (!it2.hasNext() || !result) { return false; } result = compareJavaRowAndOrcStruct(((AvroRow) it1.next()).getRow(), (OrcStruct) it2.next()); } return result; }); } /** * Given the fact that we couldn't access OrcStruct easily, here uses the hacky way(reflection) * to go around access modifier for integration test purpose only. * @param realRow A row containing a list of Java objects. * @param struct An {@link OrcStruct} which essentially is a list of {@link Writable} objects. */ private boolean compareJavaRowAndOrcStruct(Object realRow, OrcStruct struct) { boolean isIdentical = true; ArrayList<Object> javaObjRow = (ArrayList) realRow; try { Field objectArr = OrcStruct.class.getDeclaredField("fields"); objectArr.setAccessible(true); Object[] dataArr = (Object[]) objectArr.get(struct); int index = 0; for (Object dataField : dataArr) { if (dataField instanceof OrcStruct) { isIdentical = isIdentical && compareJavaRowAndOrcStruct(javaObjRow.get(index), (OrcStruct) dataField); } else { isIdentical = isIdentical && objCastHelper(javaObjRow.get(index), (Writable) dataField); } index++; } } catch (NoSuchFieldException | IllegalAccessException nfe) { throw new RuntimeException("Failed in compare a java object row and orcstruct"); } return isIdentical; } /** * All Writable objects passed in here are guaranteed to be primitive writable objects. */ private boolean objCastHelper(Object javaObj, Writable obj) { if (obj instanceof IntWritable) { return ((IntWritable) obj).get() == (Integer) javaObj; } else if (obj instanceof Text) { return (obj).toString().equals(javaObj); } else if (obj instanceof LongWritable) { return ((LongWritable) obj).get() == (Long) javaObj; } else if (obj instanceof ShortWritable) { return ((ShortWritable) obj).get() == (Short) javaObj; } else if (obj instanceof DoubleWritable) { return ((DoubleWritable) obj).get() == (Double) javaObj; } else { throw new RuntimeException("Cannot recognize the writable type, please enrich the castHelper function"); } } /** * Materialize records in a classpath package into HDFS ORC records. * @param baseResource name of the package. The package should contain the following: * - Exactly one resource called orcSchema containing the schema of the records * (or an explicit schema passed as an argument). * - One or more data files called *.json containing the records. * Note that .avsc will not be used in Orc related operation. * * @param fs * @param targetPath the path where the records will be written. * @param schema * @return */ @Override public TypeInfo writeJsonResourceRecordsAsBinary(String baseResource, @Nullable FileSystem fs, Path targetPath, @Nullable TypeInfo schema) throws IOException { TreeMap<String, OrcRowIterator> recordMap = readAllRecordsInJsonResource(baseResource, schema); TypeInfo outputSchema = recordMap.lastEntry().getValue().getTypeInfo(); for (Map.Entry<String, OrcRowIterator> entry : recordMap.entrySet()) { writeAsOrcBinary(entry.getValue(), outputSchema, new Path(targetPath, removeExtension(entry.getKey()))); } return outputSchema; } /** * AvroRow version of writeAsOrcBinary */ private void writeAsOrcBinary(OrcRowIterator input, TypeInfo schema, Path outputPath) throws IOException { Configuration configuration = new Configuration(); // Note that it doesn't support schema evolution at all. // If the schema in realRow is inconsistent with given schema, writing into disk // would run into failure. ObjectInspector oi = TypeInfoUtils.getStandardWritableObjectInspectorFromTypeInfo(schema); OrcFile.WriterOptions options = OrcFile.writerOptions(configuration).inspector(oi); Writer writer = null; while (input.hasNext()) { AvroRow avroRow = (AvroRow) input.next(); if (writer == null) { options.inspector(avroRow.getInspector()); writer = OrcFile.createWriter(outputPath, options); } writer.addRow(avroRow.realRow); } if (writer != null) { writer.close(); } } // ORC-File Reading related functions // There's no GenericRecord for ORC existed(so that OrcStruct even doesn't provide readFields as // it is responsible to transform a Writable into GenericRecord in Avro world. @Override public TreeMap<String, OrcRowIterator> readAllRecordsInJsonResource(String baseResource, @Nullable TypeInfo schema) throws IOException { TypeInfo orcSchema; try { if (schema == null) { File schemaFile = new File(baseResource, "schema.avsc"); String schemaResource = schemaFile.toString(); orcSchema = convertAvroSchemaToOrcSchema(readAvscSchema(schemaResource, OrcTestTools.class)); } else { orcSchema = schema; } } catch (SerDeException se) { throw new RuntimeException("Provided Avro Schema cannot be transformed to ORC schema", se); } TreeMap<String, OrcRowIterator> output = new TreeMap<>(); for (String file : getJsonFileSetByResourceRootName(baseResource)) { log.info("Reading json record from " + file); String name = PathUtils.relativizePath(new Path(file), new Path(baseResource)).toString(); output.put(name, readRecordsFromJsonInputStream(orcSchema, file)); } return output; } public static class AvroRow implements Writable { Object realRow; ObjectInspector inspector; public AvroRow(Object row, ObjectInspector inspector) { this.realRow = row; this.inspector = inspector; } @Override public void write(DataOutput dataOutput) throws IOException { throw new UnsupportedOperationException("can't write the bundle"); } @Override public void readFields(DataInput dataInput) throws IOException { throw new UnsupportedOperationException("can't read the bundle"); } ObjectInspector getInspector() { return inspector; } Object getRow() { return realRow; } } /** * Deserialize json object into a list of java object as a row, and transform each of java object * into {@link Writable} counterpart for constructing {@link OrcStruct}, in convenience of Orc reading and writing. * * @param typeInfo The ORC schema in {@link TypeInfo} format. * @param file The file name in String format. * @return */ private OrcRowIterator readRecordsFromJsonInputStream(TypeInfo typeInfo, String file) throws IOException { InputStream is = OrcTestTools.class.getClassLoader().getResourceAsStream(file); // This getParent.getParent is dirty due to we need to simulate multiple-partitions scenarios in iTest. String schemaResourceName = new File(new File(file).getParentFile().getParent(), "schema.avsc").toString(); Schema attemptedSchema = readAvscSchema(schemaResourceName, OrcTestTools.class); final Schema avroSchema = attemptedSchema == null ? readAvscSchema(new File(new File(file).getParent(), "schema.avsc").toString(), OrcTestTools.class) : attemptedSchema; GenericDatumReader<GenericRecord> reader = new GenericDatumReader<>(avroSchema); Decoder decoder = DecoderFactory.get().jsonDecoder(avroSchema, is); return new OrcRowIterator(typeInfo, new AbstractIterator<Writable>() { @Override protected Writable computeNext() { try { GenericRecord record = reader.read(null, decoder); return getAvroWritable(record, avroSchema); } catch (IOException e) { try { is.close(); } catch (IOException ioec) { log.warn("Failed to read record from inputstream, will close it immediately", ioec); } endOfData(); return null; } } }); } /** * From each record, transformed to {@link AvroRow} object for writing. * One can also choose to use OrcSerDe to obtain ORC-associated writable object. * * Using return object of this method would enable a self-maintained ORC writer(not from OrcOutputFormat) * to write object. */ private Writable getAvroWritable(GenericRecord record, Schema avroSchema) { try { // Construct AvroSerDe with proper schema and deserialize into Hive object. AvroSerDe serDe = new AvroSerDe(); Properties propertiesWithSchema = new Properties(); propertiesWithSchema.setProperty(AvroSerdeUtils.AvroTableProperties.SCHEMA_LITERAL.getPropName(), avroSchema.toString()); serDe.initialize(null, propertiesWithSchema); AvroGenericRecordWritable avroGenericRecordWritable = new AvroGenericRecordWritable(record); avroGenericRecordWritable.setFileSchema(avroSchema); Object avroDeserialized = serDe.deserialize(avroGenericRecordWritable); ObjectInspector avroOI = new AvroObjectInspectorGenerator(avroSchema).getObjectInspector(); return new AvroRow(avroDeserialized, avroOI); } catch (SerDeException se) { throw new RuntimeException("Failed in SerDe exception:", se); } } /** * Reading ORC file into in-memory representation. */ @Override public TreeMap<String, OrcRowIterator> readAllRecordsInBinaryDirectory(FileSystem fs, Path path) throws IOException { TreeMap<String, OrcRowIterator> output = new TreeMap<>(); if (!fs.exists(path)) { return output; } PathFilter pathFilter = new HiddenFilter(); for (FileStatus status : FileListUtils.listFilesRecursively(fs, path, pathFilter)) { String key = PathUtils.relativizePath(status.getPath(), path).toString(); Reader orcReader = OrcFile.createReader(fs, status.getPath()); RecordReader recordReader = orcReader.rows(); output.put(key, new OrcRowIterator(TypeInfoUtils.getTypeInfoFromObjectInspector(orcReader.getObjectInspector()), new AbstractIterator<Writable>() { @Override protected Writable computeNext() { try { if (recordReader.hasNext()) { return (Writable) recordReader.next(null); } else { recordReader.close(); endOfData(); return null; } } catch (IOException ioe) { log.warn("Failed to process orc record reader, will terminate reader immediately", ioe); endOfData(); return null; } } })); } return output; } /** * An iterator over {@link OrcStruct} which is also aware of schema( Represented in {@link TypeInfo}). */ @AllArgsConstructor public static class OrcRowIterator implements Iterator<Writable> { @Getter private final TypeInfo typeInfo; @Delegate private final Iterator<Writable> it; } /** * Convert Avro schema into TypeInfo. * Current version of Hive used by Gobblin open-source(1.0.1) doesn't have {@link org.apache.orc.TypeDescription} * and utilities associated with it. So instead {@link TypeInfo} is being used to represent Orc schema. * Note that {@link TypeInfo} is not case preserving as it is actually the internal schema representation of Hive. */ public static TypeInfo convertAvroSchemaToOrcSchema(Schema avroSchema) throws SerDeException { return TypeInfoUtils.getTypeInfoFromObjectInspector( new AvroObjectInspectorGenerator(avroSchema).getObjectInspector()); } }
3,931
0
Create_ds/gobblin/gobblin-binary-management/src/main/java/org/apache/gobblin
Create_ds/gobblin/gobblin-binary-management/src/main/java/org/apache/gobblin/binary_creation/DataTestTools.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.binary_creation; import java.io.IOException; import java.util.Collection; import java.util.Iterator; import java.util.TreeMap; import java.util.function.BiFunction; import lombok.extern.slf4j.Slf4j; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; /** * A ToolKit that will be used for: * - Creating binary-format file(Avro, ORC) using records declared in txt(.json) file, and schema defined in .avsc file. * - Deserializing binary-format file into traversable in-memory objects. * - Verifying if contents in two binary-format file are identical with certain constraints. * * * @param <T> Iterator containing specific type of a record row, * e.g. {@link org.apache.avro.generic.GenericRecord} for Avro. * @param <S> Schema type of a specific data format. */ @Slf4j public abstract class DataTestTools<T, S> { /** * Verify that the two inputs contain the same records in the same file names. Any fields listed in * blacklistRecordFields will not be used for comparison. * Note that this method is destructive to the input records. * @param expected Expected records map, keyed by file name. * @param observed Observed records map, keyed by file name * @param allowDifferentOrder True if allowing fields arranged in different order in comparison of two records. * @param blacklistRecordFields Configurable set of fields that won't be included for comparison of two records. * @param allowDifferentSchema True if schema info (for avro, schema can contain attributes which is not necessary * to be included for comparison) * @return */ public abstract boolean checkSameFilesAndRecords(TreeMap<String, T> expected, TreeMap<String, T> observed, boolean allowDifferentOrder, Collection<String> blacklistRecordFields, boolean allowDifferentSchema); /** * Write a resource file under a certain path as specified binary format file, like Avro, ORC. * @param baseResource Resource folder that contain JSON files. * @param fs * @param targetPath Output Path. * @param schema The schema of outputed binary file * @return * @throws IOException */ public abstract S writeJsonResourceRecordsAsBinary(String baseResource, FileSystem fs, Path targetPath, S schema) throws IOException; /** * Read all records in a json base resource in classpath into a map from file name to iterator of T object. * @param baseResource Base path of the resource directory that contains json file. * @param schema The schema of records. * @return A map between file name to an iterator of objects contained in path. */ public abstract TreeMap<String, T> readAllRecordsInJsonResource(String baseResource, S schema) throws IOException; /** * Read binary-format records into a map from file name to an iterator of T object. * @param fs File system object. * @param path File path * @return A map between file name to an iterator of objects contained in path. * @throws IOException */ public abstract TreeMap<String, T> readAllRecordsInBinaryDirectory(FileSystem fs, Path path) throws IOException; /** * Compare two iterators in T type. */ <T> boolean compareIterators(Iterator<T> expected, Iterator<T> observed, BiFunction<T, T, Boolean> comparator) { while (expected.hasNext()) { if (!observed.hasNext()) { log.error("Expected has more elements than observed."); return false; } T t1 = expected.next(); T t2 = observed.next(); boolean equals = comparator == null ? t1.equals(t2) : comparator.apply(t1, t2); if (!equals) { log.error(String.format("Mismatch: %s does not equal %s.", t1, t2)); return false; } } if (observed.hasNext()) { log.error("Observed has more elements than expected."); return false; } return true; } }
3,932
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/GobblinTemporalConfigurationKeys.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.temporal.workflows.helloworld.HelloWorldJobLauncher; import org.apache.gobblin.temporal.workflows.helloworld.HelloWorldWorker; /** * A central place for configuration related constants of a Gobblin Temporal. */ @Alpha public interface GobblinTemporalConfigurationKeys { String PREFIX = "gobblin.temporal."; String WORKER_CLASS = PREFIX + "worker.class"; String DEFAULT_WORKER_CLASS = HelloWorldWorker.class.getName(); String GOBBLIN_TEMPORAL_NAMESPACE = PREFIX + "namespace"; String DEFAULT_GOBBLIN_TEMPORAL_NAMESPACE = PREFIX + "namespace"; String GOBBLIN_TEMPORAL_TASK_QUEUE = PREFIX + "task.queue.name"; String DEFAULT_GOBBLIN_TEMPORAL_TASK_QUEUE = "GobblinTemporalTaskQueue"; String GOBBLIN_TEMPORAL_JOB_LAUNCHER_PREFIX = PREFIX + "job.launcher."; String GOBBLIN_TEMPORAL_JOB_LAUNCHER_CLASS = GOBBLIN_TEMPORAL_JOB_LAUNCHER_PREFIX + "class"; String DEFAULT_GOBBLIN_TEMPORAL_JOB_LAUNCHER_CLASS = HelloWorldJobLauncher.class.getName(); String GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_PREFIX = GOBBLIN_TEMPORAL_JOB_LAUNCHER_PREFIX + "arg."; /** * Number of worker processes to spin up per task runner * NOTE: If this size is too large, your container can OOM and halt execution unexpectedly. It's recommended not to touch * this parameter */ String TEMPORAL_NUM_WORKERS_PER_CONTAINER = PREFIX + "num.workers.per.container"; int DEFAULT_TEMPORAL_NUM_WORKERS_PER_CONTAINERS = 1; String TEMPORAL_CONNECTION_STRING = PREFIX + "connection.string"; }
3,933
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/cluster/AbstractTemporalWorker.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.cluster; import java.util.Arrays; import com.typesafe.config.Config; import io.temporal.client.WorkflowClient; import io.temporal.worker.Worker; import io.temporal.worker.WorkerFactory; import io.temporal.worker.WorkerOptions; import org.apache.gobblin.temporal.GobblinTemporalConfigurationKeys; import org.apache.gobblin.util.ConfigUtils; /** Basic boilerplate for a {@link TemporalWorker} to register its activity and workflow capabilities and listen on a particular queue */ public abstract class AbstractTemporalWorker implements TemporalWorker { private final WorkflowClient workflowClient; private final String queueName; private final WorkerFactory workerFactory; private final Config config; public AbstractTemporalWorker(Config cfg, WorkflowClient client) { config = cfg; workflowClient = client; queueName = ConfigUtils.getString(cfg, GobblinTemporalConfigurationKeys.GOBBLIN_TEMPORAL_TASK_QUEUE, GobblinTemporalConfigurationKeys.DEFAULT_GOBBLIN_TEMPORAL_TASK_QUEUE); // Create a Worker factory that can be used to create Workers that poll specific Task Queues. workerFactory = WorkerFactory.newInstance(workflowClient); stashWorkerConfig(cfg); } @Override public void start() { Worker worker = workerFactory.newWorker(queueName, createWorkerOptions()); // This Worker hosts both Workflow and Activity implementations. // Workflows are stateful, so you need to supply a type to create instances. worker.registerWorkflowImplementationTypes(getWorkflowImplClasses()); // Activities are stateless and thread safe, so a shared instance is used. worker.registerActivitiesImplementations(getActivityImplInstances()); // Start polling the Task Queue. workerFactory.start(); } @Override public void shutdown() { workerFactory.shutdown(); } protected WorkerOptions createWorkerOptions() { return null; } /** @return workflow types for *implementation* classes (not interface) */ protected abstract Class<?>[] getWorkflowImplClasses(); /** @return activity instances; NOTE: activities must be stateless and thread-safe, so a shared instance is used. */ protected abstract Object[] getActivityImplInstances(); private final void stashWorkerConfig(Config cfg) { // stash in association with... WorkerConfig.forWorker(this.getClass(), cfg); // the worker itself Arrays.stream(getWorkflowImplClasses()).forEach(clazz -> WorkerConfig.withImpl(clazz, cfg)); // its workflow impls Arrays.stream(getActivityImplInstances()).forEach(obj -> WorkerConfig.withImpl(obj.getClass(), cfg)); // its activity impls } }
3,934
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/cluster/GobblinTemporalTaskRunner.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.cluster; import java.io.IOException; import java.net.InetAddress; import java.net.UnknownHostException; import java.nio.file.Paths; import java.util.ArrayList; import java.util.Collection; import java.util.List; import java.util.Properties; import java.util.UUID; import org.apache.commons.cli.CommandLine; import org.apache.commons.cli.DefaultParser; import org.apache.commons.cli.HelpFormatter; import org.apache.commons.cli.Option; import org.apache.commons.cli.Options; import org.apache.commons.cli.ParseException; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.slf4j.Logger; import org.slf4j.LoggerFactory; import com.google.api.client.repackaged.com.google.common.annotations.VisibleForTesting; import com.google.common.base.Optional; import com.google.common.eventbus.EventBus; import com.google.common.eventbus.Subscribe; import com.google.common.util.concurrent.Service; import com.typesafe.config.Config; import com.typesafe.config.ConfigFactory; import com.typesafe.config.ConfigValueFactory; import io.temporal.client.WorkflowClient; import io.temporal.serviceclient.WorkflowServiceStubs; import lombok.Getter; import lombok.Setter; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.cluster.ContainerHealthCheckException; import org.apache.gobblin.cluster.ContainerHealthMetricsService; import org.apache.gobblin.cluster.ContainerMetrics; import org.apache.gobblin.cluster.GobblinClusterConfigurationKeys; import org.apache.gobblin.cluster.GobblinClusterManager; import org.apache.gobblin.cluster.GobblinClusterUtils; import org.apache.gobblin.cluster.TaskRunnerSuiteBase; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.configuration.State; import org.apache.gobblin.instrumented.StandardMetricsBridge; import org.apache.gobblin.metrics.GobblinMetrics; import org.apache.gobblin.metrics.MultiReporterException; import org.apache.gobblin.metrics.RootMetricContext; import org.apache.gobblin.metrics.event.EventSubmitter; import org.apache.gobblin.metrics.event.GobblinEventBuilder; import org.apache.gobblin.metrics.reporter.util.MetricReportUtils; import org.apache.gobblin.runtime.api.TaskEventMetadataGenerator; import org.apache.gobblin.temporal.GobblinTemporalConfigurationKeys; import org.apache.gobblin.temporal.workflows.client.TemporalWorkflowClientFactory; import org.apache.gobblin.util.ClassAliasResolver; import org.apache.gobblin.util.ConfigUtils; import org.apache.gobblin.util.FileUtils; import org.apache.gobblin.util.HadoopUtils; import org.apache.gobblin.util.JvmUtils; import org.apache.gobblin.util.TaskEventMetadataUtils; import org.apache.gobblin.util.event.ContainerHealthCheckFailureEvent; import org.apache.gobblin.util.reflection.GobblinConstructorUtils; /** * The main class running in the containers managing services for running Gobblin * {@link org.apache.gobblin.source.workunit.WorkUnit}s. * * <p> * If for some reason, the container exits or gets killed, the {@link GobblinClusterManager} will * be notified for the completion of the container and will start a new container to replace this one. * </p> * */ @Alpha public class GobblinTemporalTaskRunner implements StandardMetricsBridge { // Working directory key for applications. This config is set dynamically. public static final String CLUSTER_APP_WORK_DIR = GobblinClusterConfigurationKeys.GOBBLIN_CLUSTER_PREFIX + "appWorkDir"; private static final Logger logger = LoggerFactory.getLogger(GobblinTemporalTaskRunner.class); static final java.nio.file.Path CLUSTER_CONF_PATH = Paths.get("generated-gobblin-cluster.conf"); private final Optional<ContainerMetrics> containerMetrics; private final Path appWorkPath; private boolean isTaskDriver; @Getter private volatile boolean started = false; private volatile boolean stopInProgress = false; private volatile boolean isStopped = false; @Getter @Setter private volatile boolean healthCheckFailed = false; protected final String taskRunnerId; protected final EventBus eventBus = new EventBus(GobblinTemporalTaskRunner.class.getSimpleName()); protected final Config clusterConfig; @Getter protected final FileSystem fs; protected final String applicationName; protected final String applicationId; protected final int numTemporalWorkers; protected final String temporalQueueName; private final boolean isMetricReportingFailureFatal; private final boolean isEventReportingFailureFatal; private final List<TemporalWorker> workers; public GobblinTemporalTaskRunner(String applicationName, String applicationId, String taskRunnerId, Config config, Optional<Path> appWorkDirOptional) throws Exception { GobblinClusterUtils.setSystemProperties(config); //Add dynamic config config = GobblinClusterUtils.addDynamicConfig(config); this.isTaskDriver = ConfigUtils.getBoolean(config, GobblinClusterConfigurationKeys.TASK_DRIVER_ENABLED,false); this.taskRunnerId = taskRunnerId; this.applicationName = applicationName; this.applicationId = applicationId; Configuration conf = HadoopUtils.newConfiguration(); this.fs = GobblinClusterUtils.buildFileSystem(config, conf); this.appWorkPath = initAppWorkDir(config, appWorkDirOptional); this.clusterConfig = saveConfigToFile(config); logger.info("Configured GobblinTaskRunner work dir to: {}", this.appWorkPath.toString()); this.containerMetrics = buildContainerMetrics(); this.numTemporalWorkers = ConfigUtils.getInt(config, GobblinTemporalConfigurationKeys.TEMPORAL_NUM_WORKERS_PER_CONTAINER, GobblinTemporalConfigurationKeys.DEFAULT_TEMPORAL_NUM_WORKERS_PER_CONTAINERS); this.temporalQueueName = ConfigUtils.getString(config, GobblinTemporalConfigurationKeys.GOBBLIN_TEMPORAL_TASK_QUEUE, GobblinTemporalConfigurationKeys.DEFAULT_GOBBLIN_TEMPORAL_TASK_QUEUE); this.isMetricReportingFailureFatal = ConfigUtils.getBoolean(this.clusterConfig, ConfigurationKeys.GOBBLIN_TASK_METRIC_REPORTING_FAILURE_FATAL, ConfigurationKeys.DEFAULT_GOBBLIN_TASK_METRIC_REPORTING_FAILURE_FATAL); this.isEventReportingFailureFatal = ConfigUtils.getBoolean(this.clusterConfig, ConfigurationKeys.GOBBLIN_TASK_EVENT_REPORTING_FAILURE_FATAL, ConfigurationKeys.DEFAULT_GOBBLIN_TASK_EVENT_REPORTING_FAILURE_FATAL); this.workers = new ArrayList<>(); logger.info("GobblinTaskRunner({}): applicationName {}, applicationId {}, taskRunnerId {}, config {}, appWorkDir {}", this.isTaskDriver ? "taskDriver" : "worker", applicationName, applicationId, taskRunnerId, config, appWorkDirOptional); } public TaskRunnerSuiteBase.Builder getTaskRunnerSuiteBuilder() throws ReflectiveOperationException { String builderStr = ConfigUtils.getString(this.clusterConfig, GobblinClusterConfigurationKeys.TASK_RUNNER_SUITE_BUILDER, TaskRunnerSuiteBase.Builder.class.getName()); String hostName = ""; try { hostName = InetAddress.getLocalHost().getHostName(); } catch (UnknownHostException e) { logger.warn("Cannot find host name for Helix instance: {}"); } TaskRunnerSuiteBase.Builder builder = GobblinConstructorUtils.<TaskRunnerSuiteBase.Builder>invokeLongestConstructor( new ClassAliasResolver(TaskRunnerSuiteBase.Builder.class) .resolveClass(builderStr), this.clusterConfig); return builder.setAppWorkPath(this.appWorkPath) .setContainerMetrics(this.containerMetrics) .setFileSystem(this.fs) .setApplicationId(applicationId) .setApplicationName(applicationName) .setContainerId(taskRunnerId) .setHostName(hostName); } private Path initAppWorkDir(Config config, Optional<Path> appWorkDirOptional) { return appWorkDirOptional.isPresent() ? appWorkDirOptional.get() : GobblinClusterUtils .getAppWorkDirPathFromConfig(config, this.fs, this.applicationName, this.applicationId); } private Config saveConfigToFile(Config config) throws IOException { Config newConf = config .withValue(CLUSTER_APP_WORK_DIR, ConfigValueFactory.fromAnyRef(this.appWorkPath.toString())); ConfigUtils configUtils = new ConfigUtils(new FileUtils()); configUtils.saveConfigToFile(newConf, CLUSTER_CONF_PATH); return newConf; } /** * Start this {@link GobblinTemporalTaskRunner} instance. */ public void start() throws ContainerHealthCheckException { logger.info("Calling start method in GobblinTemporalTaskRunner"); logger.info(String.format("Starting in container %s", this.taskRunnerId)); // Start metric reporting initMetricReporter(); // Add a shutdown hook so the task scheduler gets properly shutdown addShutdownHook(); try { for (int i = 0; i < this.numTemporalWorkers; i++) { workers.add(initiateWorker()); } }catch (Exception e) { logger.info(e + " for initiate workers"); throw new RuntimeException(e); } } private TemporalWorker initiateWorker() throws Exception { logger.info("Starting Temporal Worker"); String connectionUri = clusterConfig.getString(GobblinTemporalConfigurationKeys.TEMPORAL_CONNECTION_STRING); WorkflowServiceStubs service = TemporalWorkflowClientFactory.createServiceInstance(connectionUri); String namespace = ConfigUtils.getString(clusterConfig, GobblinTemporalConfigurationKeys.GOBBLIN_TEMPORAL_NAMESPACE, GobblinTemporalConfigurationKeys.DEFAULT_GOBBLIN_TEMPORAL_NAMESPACE); WorkflowClient client = TemporalWorkflowClientFactory.createClientInstance(service, namespace); String workerClassName = ConfigUtils.getString(clusterConfig, GobblinTemporalConfigurationKeys.WORKER_CLASS, GobblinTemporalConfigurationKeys.DEFAULT_WORKER_CLASS); TemporalWorker worker = GobblinConstructorUtils.invokeLongestConstructor( (Class<TemporalWorker>)Class.forName(workerClassName), clusterConfig, client); worker.start(); logger.info("A new worker is started."); return worker; } private void initMetricReporter() { if (this.containerMetrics.isPresent()) { try { this.containerMetrics.get() .startMetricReportingWithFileSuffix(ConfigUtils.configToState(this.clusterConfig), this.taskRunnerId); } catch (MultiReporterException ex) { if (MetricReportUtils.shouldThrowException(logger, ex, this.isMetricReportingFailureFatal, this.isEventReportingFailureFatal)) { throw new RuntimeException(ex); } } } } public synchronized void stop() { if (this.isStopped) { logger.info("Gobblin Task runner is already stopped."); return; } if (this.stopInProgress) { logger.info("Gobblin Task runner stop already in progress."); return; } this.stopInProgress = true; logger.info("Stopping the Gobblin Task runner"); // Stop metric reporting if (this.containerMetrics.isPresent()) { this.containerMetrics.get().stopMetricsReporting(); } workers.forEach(TemporalWorker::shutdown); logger.info("All services are stopped."); this.isStopped = true; } /** * Creates and returns a {@link List} of additional {@link Service}s that should be run in this * {@link GobblinTemporalTaskRunner}. Sub-classes that need additional {@link Service}s to run, should override this method * * @return a {@link List} of additional {@link Service}s to run. */ protected List<Service> getServices() { List<Service> serviceList = new ArrayList<>(); if (ConfigUtils.getBoolean(this.clusterConfig, GobblinClusterConfigurationKeys.CONTAINER_HEALTH_METRICS_SERVICE_ENABLED, GobblinClusterConfigurationKeys.DEFAULT_CONTAINER_HEALTH_METRICS_SERVICE_ENABLED)) { serviceList.add(new ContainerHealthMetricsService(clusterConfig)); } return serviceList; } @VisibleForTesting boolean isStopped() { return this.isStopped; } private void addShutdownHook() { Runtime.getRuntime().addShutdownHook(new Thread() { @Override public void run() { logger.info("Running the shutdown hook"); GobblinTemporalTaskRunner.this.stop(); } }); } private Optional<ContainerMetrics> buildContainerMetrics() { Properties properties = ConfigUtils.configToProperties(this.clusterConfig); if (GobblinMetrics.isEnabled(properties)) { logger.info("Container metrics are enabled"); return Optional.of(ContainerMetrics .get(ConfigUtils.configToState(clusterConfig), this.applicationName, this.taskRunnerId)); } else { return Optional.absent(); } } // hard coded for now @Override public Collection<StandardMetrics> getStandardMetricsCollection() { return null; } @Subscribe public void handleContainerHealthCheckFailureEvent(ContainerHealthCheckFailureEvent event) { logger.error("Received {} from: {}", event.getClass().getSimpleName(), event.getClassName()); logger.error("Submitting a ContainerHealthCheckFailureEvent.."); submitEvent(event); logger.error("Stopping GobblinTaskRunner..."); GobblinTemporalTaskRunner.this.setHealthCheckFailed(true); GobblinTemporalTaskRunner.this.stop(); } private void submitEvent(ContainerHealthCheckFailureEvent event) { EventSubmitter eventSubmitter = new EventSubmitter.Builder(RootMetricContext.get(), getClass().getPackage().getName()).build(); GobblinEventBuilder eventBuilder = new GobblinEventBuilder(event.getClass().getSimpleName()); State taskState = ConfigUtils.configToState(event.getConfig()); //Add task metadata such as taskId, containerId, and workflowId if configured TaskEventMetadataGenerator taskEventMetadataGenerator = TaskEventMetadataUtils.getTaskEventMetadataGenerator(taskState); eventBuilder.addAdditionalMetadata(taskEventMetadataGenerator.getMetadata(taskState, event.getClass().getSimpleName())); eventBuilder.addAdditionalMetadata(event.getMetadata()); eventSubmitter.submit(eventBuilder); } private static String getApplicationId() { return "1"; } private static String getTaskRunnerId() { return UUID.randomUUID().toString(); } public static Options buildOptions() { Options options = new Options(); options.addOption("a", GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME, true, "Application name"); options.addOption("d", GobblinClusterConfigurationKeys.APPLICATION_ID_OPTION_NAME, true, "Application id"); options.addOption("i", GobblinClusterConfigurationKeys.HELIX_INSTANCE_NAME_OPTION_NAME, true, "Helix instance name"); options.addOption(Option.builder("t").longOpt(GobblinClusterConfigurationKeys.HELIX_INSTANCE_TAGS_OPTION_NAME) .hasArg(true).required(false).desc("Helix instance tags").build()); return options; } public static void printUsage(Options options) { HelpFormatter formatter = new HelpFormatter(); formatter.printHelp(GobblinClusterManager.class.getSimpleName(), options); } public static void main(String[] args) throws Exception { Options options = buildOptions(); try { CommandLine cmd = new DefaultParser().parse(options, args); if (!cmd.hasOption(GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME)) { printUsage(options); System.exit(1); } logger.info(JvmUtils.getJvmInputArguments()); String applicationName = cmd.getOptionValue(GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME); GobblinTemporalTaskRunner gobblinWorkUnitRunner = new GobblinTemporalTaskRunner(applicationName, getApplicationId(), getTaskRunnerId(), ConfigFactory.load(), Optional.<Path>absent()); gobblinWorkUnitRunner.start(); } catch (ParseException pe) { printUsage(options); System.exit(1); } } }
3,935
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/cluster/TemporalWorker.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.cluster; /** Marker interface for a temporal.io "worker", with capability to `start()` and `shutdown()` */ public interface TemporalWorker { /** Starts the worker */ void start(); /** Shuts down the worker */ void shutdown(); }
3,936
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/cluster/GobblinTemporalClusterManager.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.cluster; import java.io.IOException; import java.util.ArrayList; import java.util.Collection; import java.util.List; import java.util.Properties; import java.util.concurrent.atomic.AtomicBoolean; import org.apache.commons.cli.CommandLine; import org.apache.commons.cli.DefaultParser; import org.apache.commons.cli.HelpFormatter; import org.apache.commons.cli.Options; import org.apache.commons.cli.ParseException; import org.apache.commons.lang.StringUtils; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.slf4j.Logger; import org.slf4j.LoggerFactory; import com.google.common.base.Optional; import com.google.common.collect.ImmutableList; import com.google.common.collect.ImmutableMap; import com.google.common.eventbus.EventBus; import com.google.common.eventbus.Subscribe; import com.google.common.util.concurrent.Service; import com.typesafe.config.Config; import com.typesafe.config.ConfigFactory; import com.typesafe.config.ConfigValueFactory; import lombok.AccessLevel; import lombok.Getter; import lombok.Setter; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.cluster.ContainerHealthMetricsService; import org.apache.gobblin.cluster.GobblinClusterConfigurationKeys; import org.apache.gobblin.cluster.GobblinClusterMetricTagNames; import org.apache.gobblin.cluster.GobblinClusterUtils; import org.apache.gobblin.cluster.GobblinTaskRunner; import org.apache.gobblin.cluster.JobConfigurationManager; import org.apache.gobblin.cluster.LeadershipChangeAwareComponent; import org.apache.gobblin.cluster.event.ClusterManagerShutdownRequest; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.instrumented.StandardMetricsBridge; import org.apache.gobblin.metrics.Tag; import org.apache.gobblin.runtime.api.MutableJobCatalog; import org.apache.gobblin.runtime.app.ApplicationException; import org.apache.gobblin.runtime.app.ApplicationLauncher; import org.apache.gobblin.runtime.app.ServiceBasedAppLauncher; import org.apache.gobblin.scheduler.SchedulerService; import org.apache.gobblin.temporal.joblauncher.GobblinTemporalJobScheduler; import org.apache.gobblin.util.ConfigUtils; import org.apache.gobblin.util.JvmUtils; import org.apache.gobblin.util.reflection.GobblinConstructorUtils; /** * The central cluster manager for Gobblin Clusters. */ @Alpha @Slf4j public class GobblinTemporalClusterManager implements ApplicationLauncher, StandardMetricsBridge, LeadershipChangeAwareComponent { private static final Logger LOGGER = LoggerFactory.getLogger(GobblinTemporalClusterManager.class); private StopStatus stopStatus = new StopStatus(false); protected ServiceBasedAppLauncher applicationLauncher; // An EventBus used for communications between services running in the ApplicationMaster @Getter(AccessLevel.PUBLIC) protected final EventBus eventBus = new EventBus(GobblinTemporalClusterManager.class.getSimpleName()); protected final Path appWorkDir; @Getter protected final FileSystem fs; protected final String applicationId; // thread used to keep process up for an idle controller private Thread idleProcessThread; // set to true to stop the idle process thread private volatile boolean stopIdleProcessThread = false; private final boolean isStandaloneMode; @Getter private MutableJobCatalog jobCatalog; @Getter private JobConfigurationManager jobConfigurationManager; @Getter private GobblinTemporalJobScheduler gobblinTemporalJobScheduler; @Getter private volatile boolean started = false; protected final String clusterName; @Getter protected final Config config; public GobblinTemporalClusterManager(String clusterName, String applicationId, Config sysConfig, Optional<Path> appWorkDirOptional) throws Exception { // Set system properties passed in via application config. // overrides such as sessionTimeout. In this case, the overrides specified GobblinClusterUtils.setSystemProperties(sysConfig); //Add dynamic config this.config = GobblinClusterUtils.addDynamicConfig(sysConfig); this.clusterName = clusterName; this.isStandaloneMode = ConfigUtils.getBoolean(this.config, GobblinClusterConfigurationKeys.STANDALONE_CLUSTER_MODE_KEY, GobblinClusterConfigurationKeys.DEFAULT_STANDALONE_CLUSTER_MODE); this.applicationId = applicationId; this.fs = GobblinClusterUtils.buildFileSystem(this.config, new Configuration()); this.appWorkDir = appWorkDirOptional.isPresent() ? appWorkDirOptional.get() : GobblinClusterUtils.getAppWorkDirPathFromConfig(this.config, this.fs, clusterName, applicationId); LOGGER.info("Configured GobblinTemporalClusterManager work dir to: {}", this.appWorkDir); initializeAppLauncherAndServices(); } /** * Create the service based application launcher and other associated services * @throws Exception */ private void initializeAppLauncherAndServices() throws Exception { // Done to preserve backwards compatibility with the previously hard-coded timeout of 5 minutes Properties properties = ConfigUtils.configToProperties(this.config); if (!properties.contains(ServiceBasedAppLauncher.APP_STOP_TIME_SECONDS)) { properties.setProperty(ServiceBasedAppLauncher.APP_STOP_TIME_SECONDS, Long.toString(300)); } this.applicationLauncher = new ServiceBasedAppLauncher(properties, this.clusterName); // create a job catalog for keeping track of received jobs if a job config path is specified if (this.config.hasPath(GobblinClusterConfigurationKeys.GOBBLIN_CLUSTER_PREFIX + ConfigurationKeys.JOB_CONFIG_FILE_GENERAL_PATH_KEY)) { String jobCatalogClassName = ConfigUtils.getString(config, GobblinClusterConfigurationKeys.JOB_CATALOG_KEY, GobblinClusterConfigurationKeys.DEFAULT_JOB_CATALOG); this.jobCatalog = (MutableJobCatalog) GobblinConstructorUtils.invokeFirstConstructor(Class.forName(jobCatalogClassName), ImmutableList.of(config .getConfig(StringUtils.removeEnd(GobblinClusterConfigurationKeys.GOBBLIN_CLUSTER_PREFIX, ".")) .withFallback(this.config))); } else { this.jobCatalog = null; } SchedulerService schedulerService = new SchedulerService(properties); this.applicationLauncher.addService(schedulerService); this.gobblinTemporalJobScheduler = buildGobblinTemporalJobScheduler(config, this.appWorkDir, getMetadataTags(clusterName, applicationId), schedulerService); this.applicationLauncher.addService(this.gobblinTemporalJobScheduler); this.jobConfigurationManager = buildJobConfigurationManager(config); this.applicationLauncher.addService(this.jobConfigurationManager); if (ConfigUtils.getBoolean(this.config, GobblinClusterConfigurationKeys.CONTAINER_HEALTH_METRICS_SERVICE_ENABLED, GobblinClusterConfigurationKeys.DEFAULT_CONTAINER_HEALTH_METRICS_SERVICE_ENABLED)) { this.applicationLauncher.addService(new ContainerHealthMetricsService(config)); } } /** * Start any services required by the application launcher then start the application launcher */ private void startAppLauncherAndServices() { // other services such as the job configuration manager have a dependency on the job catalog, so it has be be // started first if (this.jobCatalog instanceof Service) { ((Service) this.jobCatalog).startAsync().awaitRunning(); } this.applicationLauncher.start(); } /** * Stop the application launcher then any services that were started outside of the application launcher */ private void stopAppLauncherAndServices() { try { this.applicationLauncher.stop(); } catch (ApplicationException ae) { LOGGER.error("Error while stopping Gobblin Cluster application launcher", ae); } if (this.jobCatalog instanceof Service) { ((Service) this.jobCatalog).stopAsync().awaitTerminated(); } } /** * Start the Gobblin Temporal Cluster Manager. */ @Override public void start() { // temporal workflow LOGGER.info("Starting the Gobblin Temporal Cluster Manager"); this.eventBus.register(this); startAppLauncherAndServices(); this.started = true; } /** * Stop the Gobblin Cluster Manager. */ @Override public synchronized void stop() { if (this.stopStatus.isStopInProgress()) { return; } this.stopStatus.setStopInprogress(true); LOGGER.info("Stopping the Gobblin Cluster Manager"); if (this.idleProcessThread != null) { try { this.idleProcessThread.join(); } catch (InterruptedException ie) { Thread.currentThread().interrupt(); } } stopAppLauncherAndServices(); } private GobblinTemporalJobScheduler buildGobblinTemporalJobScheduler(Config sysConfig, Path appWorkDir, List<? extends Tag<?>> metadataTags, SchedulerService schedulerService) throws Exception { return new GobblinTemporalJobScheduler(sysConfig, this.eventBus, appWorkDir, metadataTags, schedulerService); } private List<? extends Tag<?>> getMetadataTags(String applicationName, String applicationId) { return Tag.fromMap( new ImmutableMap.Builder<String, Object>().put(GobblinClusterMetricTagNames.APPLICATION_NAME, applicationName) .put(GobblinClusterMetricTagNames.APPLICATION_ID, applicationId).build()); } /** * Build the {@link JobConfigurationManager} for the Application Master. */ private JobConfigurationManager buildJobConfigurationManager(Config config) { try { List<Object> argumentList = (this.jobCatalog != null)? ImmutableList.of(this.eventBus, config, this.jobCatalog, this.fs) : ImmutableList.of(this.eventBus, config, this.fs); if (config.hasPath(GobblinClusterConfigurationKeys.JOB_CONFIGURATION_MANAGER_KEY)) { return (JobConfigurationManager) GobblinConstructorUtils.invokeLongestConstructor(Class.forName( config.getString(GobblinClusterConfigurationKeys.JOB_CONFIGURATION_MANAGER_KEY)), argumentList.toArray(new Object[argumentList.size()])); } else { return new JobConfigurationManager(this.eventBus, config); } } catch (ReflectiveOperationException e) { throw new RuntimeException(e); } } @SuppressWarnings("unused") @Subscribe public void handleApplicationMasterShutdownRequest(ClusterManagerShutdownRequest shutdownRequest) { stop(); } @Override public void close() throws IOException { this.applicationLauncher.close(); } @Override public Collection<StandardMetrics> getStandardMetricsCollection() { List<StandardMetrics> list = new ArrayList(); list.addAll(this.jobCatalog.getStandardMetricsCollection()); list.addAll(this.jobConfigurationManager.getStandardMetricsCollection()); return list; } /** * comment lifted from {@link org.apache.gobblin.cluster.GobblinClusterManager} * TODO for now the cluster id is hardcoded to 1 both here and in the {@link GobblinTaskRunner}. In the future, the * cluster id should be created by the {@link GobblinTemporalClusterManager} and passed to each {@link GobblinTaskRunner} */ private static String getApplicationId() { return "1"; } private static Options buildOptions() { Options options = new Options(); options.addOption("a", GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME, true, "Gobblin application name"); options.addOption("s", GobblinClusterConfigurationKeys.STANDALONE_CLUSTER_MODE, true, "Standalone cluster mode"); options.addOption("i", GobblinClusterConfigurationKeys.HELIX_INSTANCE_NAME_OPTION_NAME, true, "Helix instance name"); return options; } private static void printUsage(Options options) { HelpFormatter formatter = new HelpFormatter(); formatter.printHelp(GobblinTemporalClusterManager.class.getSimpleName(), options); } public static void main(String[] args) throws Exception { Options options = buildOptions(); try { CommandLine cmd = new DefaultParser().parse(options, args); if (!cmd.hasOption(GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME)) { printUsage(options); System.exit(1); } boolean isStandaloneClusterManager = false; if (cmd.hasOption(GobblinClusterConfigurationKeys.STANDALONE_CLUSTER_MODE)) { isStandaloneClusterManager = Boolean.parseBoolean(cmd.getOptionValue(GobblinClusterConfigurationKeys.STANDALONE_CLUSTER_MODE, "false")); } LOGGER.info(JvmUtils.getJvmInputArguments()); Config config = ConfigFactory.load(); if (cmd.hasOption(GobblinClusterConfigurationKeys.HELIX_INSTANCE_NAME_OPTION_NAME)) { config = config.withValue(GobblinClusterConfigurationKeys.HELIX_INSTANCE_NAME_KEY, ConfigValueFactory.fromAnyRef(cmd.getOptionValue( GobblinClusterConfigurationKeys.HELIX_INSTANCE_NAME_OPTION_NAME))); } if (isStandaloneClusterManager) { config = config.withValue(GobblinClusterConfigurationKeys.STANDALONE_CLUSTER_MODE_KEY, ConfigValueFactory.fromAnyRef(true)); } try (GobblinTemporalClusterManager GobblinTemporalClusterManager = new GobblinTemporalClusterManager( cmd.getOptionValue(GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME), getApplicationId(), config, Optional.<Path>absent())) { GobblinTemporalClusterManager.start(); } } catch (ParseException pe) { printUsage(options); System.exit(1); } } @Override public void becomeActive() { startAppLauncherAndServices(); } @Override public void becomeStandby() { stopAppLauncherAndServices(); try { initializeAppLauncherAndServices(); } catch (Exception e) { throw new RuntimeException("Exception reinitializing app launcher services ", e); } } static class StopStatus { @Getter @Setter AtomicBoolean isStopInProgress; public StopStatus(boolean inProgress) { isStopInProgress = new AtomicBoolean(inProgress); } public void setStopInprogress (boolean inProgress) { isStopInProgress.set(inProgress); } public boolean isStopInProgress () { return isStopInProgress.get(); } } }
3,937
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/cluster/WorkerConfig.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.cluster; import java.util.Optional; import java.util.concurrent.ConcurrentHashMap; import lombok.extern.slf4j.Slf4j; import com.typesafe.config.Config; /** * Static holder to stash the {@link Config} used to construct each kind of {@link org.apache.gobblin.temporal.cluster.TemporalWorker} * (within the current JVM). Lookup may be by either the {@link Class} of the worker or of any workflow or activity implementation supplied by * that worker. The objective is to facilitate sharing the worker's Config with workflow and activity implementations (running within that worker). * * ATTENTION: for sanity, construct multiple instances of the same worker always with the same {@link Config}. When this is violated, the `Config` * given to the most-recently constructed worker "wins". * * NOTE: the preservation and sharing of {@link Config} is predicated entirely on its immutability. Thank you TypeSafe! * Storage indexing uses FQ class name, not the {@link Class}, to be independent of classloader. */ @Slf4j public class WorkerConfig { private static final ConcurrentHashMap<String, Config> configByFQClassName = new ConcurrentHashMap<>(); private WorkerConfig() {} /** @return whether initialized now (vs. being previously known) */ public static boolean forWorker(Class<? extends TemporalWorker> workerClass, Config config) { return storeAs(workerClass.getName(), config); } /** @return whether initialized now (vs. being previously known) */ public static boolean withImpl(Class<?> workflowOrActivityImplClass, Config config) { return storeAs(workflowOrActivityImplClass.getName(), config); } public static Optional<Config> ofWorker(Class<? extends TemporalWorker> workerClass) { return Optional.ofNullable(configByFQClassName.get(workerClass.getName())); } public static Optional<Config> ofImpl(Class<?> workflowOrActivityImplClass) { return Optional.ofNullable(configByFQClassName.get(workflowOrActivityImplClass.getName())); } public static Optional<Config> of(Object workflowOrActivityImpl) { return ofImpl(workflowOrActivityImpl.getClass()); } private static boolean storeAs(String className, Config config) { Config prior = configByFQClassName.put(className, config); log.info("storing config of {} values as '{}'{}", config.entrySet().size(), className, prior == null ? " (new)" : ""); return prior == null; } }
3,938
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/joblauncher/GobblinTemporalJobSchedulerMetrics.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.joblauncher; import com.google.common.base.Optional; import java.util.Properties; import java.util.concurrent.ExecutorService; import java.util.concurrent.ThreadPoolExecutor; import java.util.concurrent.TimeUnit; import java.util.concurrent.atomic.AtomicLong; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.instrumented.StandardMetricsBridge; import org.apache.gobblin.metrics.ContextAwareTimer; import org.apache.gobblin.metrics.MetricContext; import org.apache.gobblin.runtime.api.JobExecutionLauncher; public class GobblinTemporalJobSchedulerMetrics extends StandardMetricsBridge.StandardMetrics { public static final String SCHEDULE_CANCELLATION_START = "scheduleCancellationStart"; public static final String SCHEDULE_CANCELLATION_END = "scheduleCancellationEnd"; public static final String TIMER_BEFORE_JOB_SCHEDULING = "timerBeforeJobScheduling"; public static final String TIMER_BEFORE_JOB_LAUNCHING = "timerBeforeJobLaunching"; public static final String TIMER_BETWEEN_JOB_SCHEDULING_AND_LAUNCHING = "timerBetwenJobSchedulingAndLaunching"; final AtomicLong numCancellationStart; final AtomicLong numCancellationComplete; final ContextAwareTimer timeBeforeJobScheduling; final ContextAwareTimer timeBeforeJobLaunching; final ContextAwareTimer timeBetwenJobSchedulingAndLaunching; final ThreadPoolExecutor threadPoolExecutor; public GobblinTemporalJobSchedulerMetrics(final ExecutorService jobExecutor, final MetricContext metricContext, int windowSizeInMin) { this.timeBeforeJobScheduling = metricContext.contextAwareTimer(TIMER_BEFORE_JOB_SCHEDULING, windowSizeInMin, TimeUnit.MINUTES); this.timeBeforeJobLaunching = metricContext.contextAwareTimer(TIMER_BEFORE_JOB_LAUNCHING, windowSizeInMin, TimeUnit.MINUTES); this.timeBetwenJobSchedulingAndLaunching = metricContext.contextAwareTimer(TIMER_BETWEEN_JOB_SCHEDULING_AND_LAUNCHING, windowSizeInMin, TimeUnit.MINUTES); this.numCancellationStart = new AtomicLong(0); this.numCancellationComplete = new AtomicLong(0); this.contextAwareMetrics.add(metricContext.newContextAwareGauge(SCHEDULE_CANCELLATION_START, ()->this.numCancellationStart.get())); this.contextAwareMetrics.add(metricContext.newContextAwareGauge(SCHEDULE_CANCELLATION_END, ()->this.numCancellationComplete.get())); this.contextAwareMetrics.add(timeBeforeJobScheduling); this.contextAwareMetrics.add(timeBeforeJobLaunching); this.contextAwareMetrics.add(timeBetwenJobSchedulingAndLaunching); this.threadPoolExecutor = (ThreadPoolExecutor) jobExecutor; // executor metrics this.contextAwareMetrics.add(metricContext.newContextAwareGauge(JobExecutionLauncher.StandardMetrics.EXECUTOR_ACTIVE_COUNT, ()->this.threadPoolExecutor.getActiveCount())); this.contextAwareMetrics.add(metricContext.newContextAwareGauge(JobExecutionLauncher.StandardMetrics.EXECUTOR_MAX_POOL_SIZE, ()->this.threadPoolExecutor.getMaximumPoolSize())); this.contextAwareMetrics.add(metricContext.newContextAwareGauge(JobExecutionLauncher.StandardMetrics.EXECUTOR_POOL_SIZE, ()->this.threadPoolExecutor.getPoolSize())); this.contextAwareMetrics.add(metricContext.newContextAwareGauge(JobExecutionLauncher.StandardMetrics.EXECUTOR_CORE_POOL_SIZE, ()->this.threadPoolExecutor.getCorePoolSize())); this.contextAwareMetrics.add(metricContext.newContextAwareGauge(JobExecutionLauncher.StandardMetrics.EXECUTOR_QUEUE_SIZE, ()->this.threadPoolExecutor.getQueue().size())); } void updateTimeBeforeJobScheduling (Properties jobProps) { long jobCreationTime = Long.parseLong(jobProps.getProperty(ConfigurationKeys.FLOW_EXECUTION_ID_KEY, "0")); Instrumented.updateTimer(Optional.of(timeBeforeJobScheduling), System.currentTimeMillis() - jobCreationTime, TimeUnit.MILLISECONDS); } void updateTimeBeforeJobLaunching (Properties jobProps) { long jobCreationTime = Long.parseLong(jobProps.getProperty(ConfigurationKeys.FLOW_EXECUTION_ID_KEY, "0")); Instrumented.updateTimer(Optional.of(timeBeforeJobLaunching), System.currentTimeMillis() - jobCreationTime, TimeUnit.MILLISECONDS); } void updateTimeBetweenJobSchedulingAndJobLaunching (long scheduledTime, long launchingTime) { Instrumented.updateTimer(Optional.of(timeBetwenJobSchedulingAndLaunching), launchingTime - scheduledTime, TimeUnit.MILLISECONDS); } }
3,939
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/joblauncher/GobblinTemporalPlanningJobLauncherMetrics.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.joblauncher; import java.util.concurrent.TimeUnit; import org.apache.gobblin.cluster.HelixJobsMapping; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.instrumented.StandardMetricsBridge; import org.apache.gobblin.metrics.ContextAwareMeter; import org.apache.gobblin.metrics.ContextAwareTimer; import org.apache.gobblin.metrics.MetricContext; public class GobblinTemporalPlanningJobLauncherMetrics extends StandardMetricsBridge.StandardMetrics { private final String metricsName; public static final String TIMER_FOR_COMPLETED_PLANNING_JOBS = "timeForCompletedPlanningJobs"; public static final String TIMER_FOR_FAILED_PLANNING_JOBS = "timeForFailedPlanningJobs"; public static final String METER_FOR_SKIPPED_PLANNING_JOBS = "skippedPlanningJobs"; final ContextAwareTimer timeForCompletedPlanningJobs; final ContextAwareTimer timeForFailedPlanningJobs; final ContextAwareMeter skippedPlanningJobs; public GobblinTemporalPlanningJobLauncherMetrics(String metricsName, final MetricContext metricContext, int windowSizeInMin, HelixJobsMapping jobsMapping) { this.metricsName = metricsName; this.timeForCompletedPlanningJobs = metricContext.contextAwareTimer(TIMER_FOR_COMPLETED_PLANNING_JOBS, windowSizeInMin, TimeUnit.MINUTES); this.timeForFailedPlanningJobs = metricContext.contextAwareTimer(TIMER_FOR_FAILED_PLANNING_JOBS, windowSizeInMin, TimeUnit.MINUTES); this.skippedPlanningJobs = metricContext.contextAwareMeter(METER_FOR_SKIPPED_PLANNING_JOBS); this.contextAwareMetrics.add(timeForCompletedPlanningJobs); this.contextAwareMetrics.add(timeForFailedPlanningJobs); } public void updateTimeForCompletedPlanningJobs(long startTime) { Instrumented.updateTimer( com.google.common.base.Optional.of(this.timeForCompletedPlanningJobs), System.currentTimeMillis() - startTime, TimeUnit.MILLISECONDS); } public void updateTimeForFailedPlanningJobs(long startTime) { Instrumented.updateTimer( com.google.common.base.Optional.of(this.timeForFailedPlanningJobs), System.currentTimeMillis() - startTime, TimeUnit.MILLISECONDS); } @Override public String getName() { return this.metricsName; } }
3,940
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/joblauncher/GobblinTemporalJobLauncher.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.joblauncher; import java.util.List; import java.util.Properties; import java.util.concurrent.ConcurrentHashMap; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import io.temporal.client.WorkflowClient; import io.temporal.serviceclient.WorkflowServiceStubs; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.cluster.GobblinClusterConfigurationKeys; import org.apache.gobblin.metrics.Tag; import org.apache.gobblin.runtime.JobLauncher; import org.apache.gobblin.source.workunit.WorkUnit; import org.apache.gobblin.temporal.cluster.GobblinTemporalTaskRunner; import static org.apache.gobblin.temporal.GobblinTemporalConfigurationKeys.*; import static org.apache.gobblin.temporal.workflows.client.TemporalWorkflowClientFactory.createClientInstance; import static org.apache.gobblin.temporal.workflows.client.TemporalWorkflowClientFactory.createServiceInstance; /** * An implementation of {@link JobLauncher} that launches a Gobblin job using the Temporal task framework. * * <p> * Each {@link WorkUnit} of the job is persisted to the {@link FileSystem} of choice and the path to the file * storing the serialized {@link WorkUnit} is passed to the Temporal task running the {@link WorkUnit} as a * user-defined property {@link GobblinClusterConfigurationKeys#WORK_UNIT_FILE_PATH}. Upon startup, the gobblin * task reads the property for the file path and de-serializes the {@link WorkUnit} from the file. * </p> * * <p> * This class is instantiated by the {@link GobblinTemporalJobScheduler} on every job submission to launch the Gobblin job. * The actual task execution happens in the {@link GobblinTemporalTaskRunner}, usually in a different process. * </p> */ @Alpha @Slf4j public abstract class GobblinTemporalJobLauncher extends GobblinJobLauncher { protected WorkflowServiceStubs workflowServiceStubs; protected WorkflowClient client; protected String queueName; public GobblinTemporalJobLauncher(Properties jobProps, Path appWorkDir, List<? extends Tag<?>> metadataTags, ConcurrentHashMap<String, Boolean> runningMap) throws Exception { super(jobProps, appWorkDir, metadataTags, runningMap); log.debug("GobblinTemporalJobLauncher: jobProps {}, appWorkDir {}", jobProps, appWorkDir); String connectionUri = jobProps.getProperty(TEMPORAL_CONNECTION_STRING); this.workflowServiceStubs = createServiceInstance(connectionUri); String namespace = jobProps.getProperty(GOBBLIN_TEMPORAL_NAMESPACE, DEFAULT_GOBBLIN_TEMPORAL_NAMESPACE); this.client = createClientInstance(workflowServiceStubs, namespace); this.queueName = jobProps.getProperty(GOBBLIN_TEMPORAL_TASK_QUEUE, DEFAULT_GOBBLIN_TEMPORAL_TASK_QUEUE); startCancellationExecutor(); } /** * Submit a job to run. */ @Override abstract protected void submitJob(List<WorkUnit> workUnits) throws Exception; @Override protected void executeCancellation() { log.info("Cancel temporal workflow"); } @Override protected void removeTasksFromCurrentJob(List<String> workUnitIdsToRemove) { log.info("Temporal removeTasksFromCurrentJob"); } protected void addTasksToCurrentJob(List<WorkUnit> workUnitsToAdd) { log.info("Temporal addTasksToCurrentJob"); } }
3,941
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/joblauncher/GobblinTemporalJobLauncherMetrics.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.joblauncher; import java.util.concurrent.TimeUnit; import org.apache.gobblin.instrumented.StandardMetricsBridge; import org.apache.gobblin.metrics.ContextAwareMeter; import org.apache.gobblin.metrics.ContextAwareTimer; import org.apache.gobblin.metrics.MetricContext; import org.apache.gobblin.runtime.api.JobExecutionLauncher; /** * Metrics that relates to jobs launched by {@link GobblinTemporalJobLauncher}. */ public class GobblinTemporalJobLauncherMetrics extends StandardMetricsBridge.StandardMetrics { private final String metricsName; final ContextAwareMeter numJobsLaunched; final ContextAwareMeter numJobsCompleted; final ContextAwareMeter numJobsCommitted; final ContextAwareMeter numJobsFailed; final ContextAwareMeter numJobsCancelled; final ContextAwareTimer timeForCompletedJobs; final ContextAwareTimer timeForFailedJobs; final ContextAwareTimer timeForCommittedJobs; public GobblinTemporalJobLauncherMetrics(String metricsName, final MetricContext metricContext, int windowSizeInMin) { this.metricsName = metricsName; this.numJobsLaunched = metricContext.contextAwareMeter(JobExecutionLauncher.StandardMetrics.NUM_JOBS_LAUNCHED); this.contextAwareMetrics.add(this.numJobsLaunched); this.numJobsCompleted = metricContext.contextAwareMeter(JobExecutionLauncher.StandardMetrics.NUM_JOBS_COMPLETED); this.contextAwareMetrics.add(this.numJobsCompleted); this.numJobsCommitted = metricContext.contextAwareMeter(JobExecutionLauncher.StandardMetrics.NUM_JOBS_COMMITTED); this.contextAwareMetrics.add(this.numJobsCommitted); this.numJobsFailed = metricContext.contextAwareMeter(JobExecutionLauncher.StandardMetrics.NUM_JOBS_FAILED); this.contextAwareMetrics.add(this.numJobsFailed); this.numJobsCancelled = metricContext.contextAwareMeter(JobExecutionLauncher.StandardMetrics.NUM_JOBS_CANCELLED); this.contextAwareMetrics.add(this.numJobsCancelled); this.contextAwareMetrics.add(metricContext.newContextAwareGauge(JobExecutionLauncher.StandardMetrics.NUM_JOBS_RUNNING, () -> (int) ( GobblinTemporalJobLauncherMetrics.this.numJobsLaunched.getCount() - GobblinTemporalJobLauncherMetrics.this.numJobsCompleted.getCount()))); this.timeForCompletedJobs = metricContext.contextAwareTimer(JobExecutionLauncher.StandardMetrics.TIMER_FOR_COMPLETED_JOBS, windowSizeInMin, TimeUnit.MINUTES); this.timeForFailedJobs = metricContext.contextAwareTimer(JobExecutionLauncher.StandardMetrics.TIMER_FOR_FAILED_JOBS, windowSizeInMin, TimeUnit.MINUTES); this.timeForCommittedJobs = metricContext.contextAwareTimer(JobExecutionLauncher.StandardMetrics.TIMER_FOR_COMMITTED_JOBS, windowSizeInMin, TimeUnit.MINUTES); this.contextAwareMetrics.add(timeForCommittedJobs); this.contextAwareMetrics.add(timeForCompletedJobs); this.contextAwareMetrics.add(timeForFailedJobs); } @Override public String getName() { return this.metricsName; } }
3,942
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/joblauncher/GobblinTemporalJobScheduler.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.joblauncher; import java.util.Collection; import java.util.List; import java.util.Properties; import java.util.concurrent.ConcurrentHashMap; import org.apache.hadoop.fs.Path; import org.slf4j.Logger; import org.slf4j.LoggerFactory; import com.google.common.collect.ImmutableList; import com.google.common.collect.Maps; import com.google.common.eventbus.EventBus; import com.google.common.eventbus.Subscribe; import com.typesafe.config.Config; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.cluster.GobblinClusterConfigurationKeys; import org.apache.gobblin.cluster.GobblinHelixJob; import org.apache.gobblin.cluster.HelixJobsMapping; import org.apache.gobblin.cluster.event.CancelJobConfigArrivalEvent; import org.apache.gobblin.cluster.event.DeleteJobConfigArrivalEvent; import org.apache.gobblin.cluster.event.NewJobConfigArrivalEvent; import org.apache.gobblin.cluster.event.UpdateJobConfigArrivalEvent; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.instrumented.StandardMetricsBridge; import org.apache.gobblin.metrics.MetricContext; import org.apache.gobblin.metrics.Tag; import org.apache.gobblin.runtime.JobException; import org.apache.gobblin.runtime.JobLauncher; import org.apache.gobblin.runtime.listeners.JobListener; import org.apache.gobblin.scheduler.JobScheduler; import org.apache.gobblin.scheduler.SchedulerService; import org.apache.gobblin.temporal.GobblinTemporalConfigurationKeys; import org.apache.gobblin.util.ConfigUtils; import org.apache.gobblin.util.PathUtils; import org.apache.gobblin.util.PropertiesUtils; import org.apache.gobblin.util.reflection.GobblinConstructorUtils; /** * An extension to {@link JobScheduler} that schedules and runs * Gobblin jobs on Temporal. * * <p> If the job should be launched from the scheduler node, * {@link GobblinTemporalJobLauncher} is invoked. * TODO(yiyang): this file should be cleaned up with HelixJobScheduler. */ @Alpha public class GobblinTemporalJobScheduler extends JobScheduler implements StandardMetricsBridge { private static final Logger LOGGER = LoggerFactory.getLogger(GobblinTemporalJobScheduler.class); private static final String COMMON_JOB_PROPS = "gobblin.common.job.props"; private final Properties commonJobProperties; private final EventBus eventBus; private final Path appWorkDir; private final List<? extends Tag<?>> metadataTags; private final ConcurrentHashMap<String, Boolean> jobRunningMap; private final MetricContext metricContext; final GobblinTemporalJobSchedulerMetrics jobSchedulerMetrics; final GobblinTemporalJobLauncherMetrics launcherMetrics; final GobblinTemporalPlanningJobLauncherMetrics planningJobLauncherMetrics; final HelixJobsMapping jobsMapping; private boolean startServicesCompleted; public GobblinTemporalJobScheduler(Config sysConfig, EventBus eventBus, Path appWorkDir, List<? extends Tag<?>> metadataTags, SchedulerService schedulerService) throws Exception { super(ConfigUtils.configToProperties(sysConfig), schedulerService); this.commonJobProperties = ConfigUtils.configToProperties(ConfigUtils.getConfigOrEmpty(sysConfig, COMMON_JOB_PROPS)); this.eventBus = eventBus; this.jobRunningMap = new ConcurrentHashMap<>(); this.appWorkDir = appWorkDir; this.metadataTags = metadataTags; this.metricContext = Instrumented.getMetricContext(new org.apache.gobblin.configuration.State(properties), this.getClass()); int metricsWindowSizeInMin = ConfigUtils.getInt(sysConfig, ConfigurationKeys.METRIC_TIMER_WINDOW_SIZE_IN_MINUTES, ConfigurationKeys.DEFAULT_METRIC_TIMER_WINDOW_SIZE_IN_MINUTES); this.launcherMetrics = new GobblinTemporalJobLauncherMetrics("launcherInScheduler", this.metricContext, metricsWindowSizeInMin); this.jobSchedulerMetrics = new GobblinTemporalJobSchedulerMetrics(this.jobExecutor, this.metricContext, metricsWindowSizeInMin); this.jobsMapping = new HelixJobsMapping(ConfigUtils.propertiesToConfig(properties), PathUtils.getRootPath(appWorkDir).toUri(), appWorkDir.toString()); this.planningJobLauncherMetrics = new GobblinTemporalPlanningJobLauncherMetrics("planningLauncherInScheduler", this.metricContext, metricsWindowSizeInMin, this.jobsMapping); this.startServicesCompleted = false; } @Override public Collection<StandardMetrics> getStandardMetricsCollection() { return ImmutableList.of(this.launcherMetrics, this.jobSchedulerMetrics, this.planningJobLauncherMetrics); } @Override protected void startUp() throws Exception { this.eventBus.register(this); super.startUp(); this.startServicesCompleted = true; } @Override public void scheduleJob(Properties jobProps, JobListener jobListener) throws JobException { try { while (!startServicesCompleted) { LOGGER.info("{} service is not fully up, waiting here...", this.getClass().getName()); Thread.sleep(1000); } scheduleJob(jobProps, jobListener, Maps.newHashMap(), GobblinHelixJob.class); } catch (Exception e) { throw new JobException("Failed to schedule job " + jobProps.getProperty(ConfigurationKeys.JOB_NAME_KEY), e); } } @Override protected void startServices() throws Exception { boolean cleanAllDistJobs = PropertiesUtils.getPropAsBoolean(this.properties, GobblinClusterConfigurationKeys.CLEAN_ALL_DIST_JOBS, String.valueOf(GobblinClusterConfigurationKeys.DEFAULT_CLEAN_ALL_DIST_JOBS)); if (cleanAllDistJobs) { for (org.apache.gobblin.configuration.State state : this.jobsMapping.getAllStates()) { String jobUri = state.getId(); LOGGER.info("Delete mapping for job " + jobUri); this.jobsMapping.deleteMapping(jobUri); } } } @Override public void runJob(Properties jobProps, JobListener jobListener) throws JobException { } @Override public GobblinTemporalJobLauncher buildJobLauncher(Properties jobProps) throws Exception { Properties combinedProps = new Properties(); combinedProps.putAll(properties); combinedProps.putAll(jobProps); Class<? extends GobblinTemporalJobLauncher> jobLauncherClass = (Class<? extends GobblinTemporalJobLauncher>) Class.forName(combinedProps.getProperty( GobblinTemporalConfigurationKeys.GOBBLIN_TEMPORAL_JOB_LAUNCHER_CLASS, GobblinTemporalConfigurationKeys.DEFAULT_GOBBLIN_TEMPORAL_JOB_LAUNCHER_CLASS)); return GobblinConstructorUtils.invokeLongestConstructor(jobLauncherClass, combinedProps, this.appWorkDir, this.metadataTags, this.jobRunningMap); } @Subscribe public void handleNewJobConfigArrival(NewJobConfigArrivalEvent newJobArrival) { String jobUri = newJobArrival.getJobName(); LOGGER.info("Received new job configuration of job " + jobUri); try { Properties jobProps = new Properties(); jobProps.putAll(this.commonJobProperties); jobProps.putAll(newJobArrival.getJobConfig()); // set uri so that we can delete this job later jobProps.setProperty(GobblinClusterConfigurationKeys.JOB_SPEC_URI, jobUri); this.jobSchedulerMetrics.updateTimeBeforeJobScheduling(jobProps); if (jobProps.containsKey(ConfigurationKeys.JOB_SCHEDULE_KEY)) { LOGGER.info("Scheduling job " + jobUri); scheduleJob(jobProps, new GobblinTemporalJobLauncherListener(this.launcherMetrics)); } else { LOGGER.info("No job schedule found, so running job " + jobUri); GobblinTemporalJobLauncherListener listener = new GobblinTemporalJobLauncherListener(this.launcherMetrics); JobLauncher launcher = buildJobLauncher(newJobArrival.getJobConfig()); launcher.launchJob(listener); } } catch (Exception je) { LOGGER.error("Failed to schedule or run job " + jobUri, je); } } @Subscribe public void handleUpdateJobConfigArrival(UpdateJobConfigArrivalEvent updateJobArrival) { LOGGER.info("Received update for job configuration of job " + updateJobArrival.getJobName()); try { handleDeleteJobConfigArrival(new DeleteJobConfigArrivalEvent(updateJobArrival.getJobName(), updateJobArrival.getJobConfig())); } catch (Exception je) { LOGGER.error("Failed to update job " + updateJobArrival.getJobName(), je); } try { handleNewJobConfigArrival(new NewJobConfigArrivalEvent(updateJobArrival.getJobName(), updateJobArrival.getJobConfig())); } catch (Exception je) { LOGGER.error("Failed to update job " + updateJobArrival.getJobName(), je); } } private void waitForJobCompletion(String jobName) { while (this.jobRunningMap.getOrDefault(jobName, false)) { LOGGER.info("Waiting for job {} to stop...", jobName); try { Thread.sleep(1000); } catch (InterruptedException e) { LOGGER.warn("Interrupted exception encountered: ", e); } } } @Subscribe public void handleDeleteJobConfigArrival(DeleteJobConfigArrivalEvent deleteJobArrival) throws InterruptedException { LOGGER.info("Received delete for job configuration of job " + deleteJobArrival.getJobName()); try { unscheduleJob(deleteJobArrival.getJobName()); } catch (JobException je) { LOGGER.error("Failed to unschedule job " + deleteJobArrival.getJobName()); } } @Subscribe public void handleCancelJobConfigArrival(CancelJobConfigArrivalEvent cancelJobArrival) throws InterruptedException { } /** * This class is responsible for running non-scheduled jobs. */ class NonScheduledJobRunner implements Runnable { private final Properties jobProps; private final GobblinTemporalJobLauncherListener jobListener; private final Long creationTimeInMillis; public NonScheduledJobRunner(Properties jobProps, GobblinTemporalJobLauncherListener jobListener) { this.jobProps = jobProps; this.jobListener = jobListener; this.creationTimeInMillis = System.currentTimeMillis(); } @Override public void run() { try { GobblinTemporalJobScheduler.this.jobSchedulerMetrics.updateTimeBeforeJobLaunching(this.jobProps); GobblinTemporalJobScheduler.this.jobSchedulerMetrics.updateTimeBetweenJobSchedulingAndJobLaunching(this.creationTimeInMillis, System.currentTimeMillis()); GobblinTemporalJobScheduler.this.runJob(this.jobProps, this.jobListener); } catch (JobException je) { LOGGER.error("Failed to run job " + this.jobProps.getProperty(ConfigurationKeys.JOB_NAME_KEY), je); } } } }
3,943
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/joblauncher/GobblinTemporalJobLauncherListener.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.joblauncher; import com.google.common.base.Optional; import java.util.concurrent.TimeUnit; import org.apache.gobblin.cluster.GobblinHelixJobLauncher; import org.apache.gobblin.instrumented.Instrumented; import org.apache.gobblin.runtime.JobContext; import org.apache.gobblin.runtime.JobState; import org.apache.gobblin.runtime.listeners.AbstractJobListener; import org.apache.gobblin.runtime.listeners.JobListener; /** * A job listener used when {@link GobblinHelixJobLauncher} launches a job. * The {@link GobblinTemporalJobLauncherMetrics} will always be passed in because * it will be be updated accordingly. */ class GobblinTemporalJobLauncherListener extends AbstractJobListener { private final GobblinTemporalJobLauncherMetrics jobLauncherMetrics; private static final String JOB_START_TIME = "jobStartTime"; GobblinTemporalJobLauncherListener(GobblinTemporalJobLauncherMetrics jobLauncherMetrics) { this.jobLauncherMetrics = jobLauncherMetrics; } @Override public void onJobPrepare(JobContext jobContext) throws Exception { super.onJobPrepare(jobContext); jobContext.getJobState().setProp(JOB_START_TIME, Long.toString(System.nanoTime())); jobLauncherMetrics.numJobsLaunched.mark(); } /** * From {@link org.apache.gobblin.runtime.AbstractJobLauncher#launchJob(JobListener)}, the final * job state should only be FAILED or COMMITTED. This means the completed jobs metrics covers * both failed jobs and committed jobs. */ @Override public void onJobCompletion(JobContext jobContext) throws Exception { super.onJobCompletion(jobContext); long startTime = jobContext.getJobState().getPropAsLong(JOB_START_TIME); jobLauncherMetrics.numJobsCompleted.mark(); Instrumented.updateTimer(Optional.of(jobLauncherMetrics.timeForCompletedJobs), System.nanoTime() - startTime, TimeUnit.NANOSECONDS); if (jobContext.getJobState().getState() == JobState.RunningState.FAILED) { jobLauncherMetrics.numJobsFailed.mark(); Instrumented.updateTimer(Optional.of(jobLauncherMetrics.timeForFailedJobs), System.nanoTime() - startTime, TimeUnit.NANOSECONDS); } else { jobLauncherMetrics.numJobsCommitted.mark(); Instrumented.updateTimer(Optional.of(jobLauncherMetrics.timeForCommittedJobs), System.nanoTime() - startTime, TimeUnit.NANOSECONDS); } } @Override public void onJobCancellation(JobContext jobContext) throws Exception { super.onJobCancellation(jobContext); jobLauncherMetrics.numJobsCancelled.mark(); } }
3,944
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/joblauncher/GobblinJobLauncher.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.joblauncher; import com.google.common.annotations.VisibleForTesting; import com.typesafe.config.Config; import com.typesafe.config.ConfigValueFactory; import java.io.IOException; import java.net.URI; import java.util.List; import java.util.Properties; import java.util.concurrent.ConcurrentHashMap; import javax.annotation.Nullable; import lombok.Getter; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.cluster.GobblinClusterConfigurationKeys; import org.apache.gobblin.cluster.GobblinClusterUtils; import org.apache.gobblin.cluster.HelixUtils; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.metrics.Tag; import org.apache.gobblin.metrics.event.CountEventBuilder; import org.apache.gobblin.metrics.event.JobEvent; import org.apache.gobblin.metrics.event.TimingEvent; import org.apache.gobblin.rest.LauncherTypeEnum; import org.apache.gobblin.runtime.AbstractJobLauncher; import org.apache.gobblin.runtime.JobException; import org.apache.gobblin.runtime.JobLauncher; import org.apache.gobblin.runtime.JobState; import org.apache.gobblin.runtime.TaskStateCollectorService; import org.apache.gobblin.runtime.listeners.JobListener; import org.apache.gobblin.runtime.util.StateStores; import org.apache.gobblin.source.workunit.WorkUnit; import org.apache.gobblin.util.ConfigUtils; import org.apache.gobblin.util.ParallelRunner; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; /** * An implementation of {@link JobLauncher} that launches a Gobblin job using the Temporal task framework. * Most of this code is lifted from {@link org.apache.gobblin.cluster.GobblinHelixJobLauncher} and maybe in the future * it may make sense to converge the code once Temporal on Gobblin is in a mature state. * * <p> * Each {@link WorkUnit} of the job is persisted to the {@link FileSystem} of choice and the path to the file * storing the serialized {@link WorkUnit} is passed to the Temporal task running the {@link WorkUnit} as a * user-defined property {@link GobblinClusterConfigurationKeys#WORK_UNIT_FILE_PATH}. Upon startup, the gobblin * task reads the property for the file path and de-serializes the {@link WorkUnit} from the file. * </p> */ @Alpha @Slf4j public abstract class GobblinJobLauncher extends AbstractJobLauncher { protected static final String WORK_UNIT_FILE_EXTENSION = ".wu"; protected final FileSystem fs; protected final Path appWorkDir; protected final Path inputWorkUnitDir; protected final Path outputTaskStateDir; // Number of ParallelRunner threads to be used for state serialization/deserialization protected final int stateSerDeRunnerThreads; protected final TaskStateCollectorService taskStateCollectorService; protected final ConcurrentHashMap<String, Boolean> runningMap; @Getter protected final StateStores stateStores; protected JobListener jobListener; protected volatile boolean jobSubmitted = false; public GobblinJobLauncher(Properties jobProps, Path appWorkDir, List<? extends Tag<?>> metadataTags, ConcurrentHashMap<String, Boolean> runningMap) throws Exception { super(jobProps, HelixUtils.initBaseEventTags(jobProps, metadataTags)); log.debug("GobblinJobLauncher: jobProps {}, appWorkDir {}", jobProps, appWorkDir); this.runningMap = runningMap; this.appWorkDir = appWorkDir; this.inputWorkUnitDir = new Path(appWorkDir, GobblinClusterConfigurationKeys.INPUT_WORK_UNIT_DIR_NAME); this.outputTaskStateDir = new Path(this.appWorkDir, GobblinClusterConfigurationKeys.OUTPUT_TASK_STATE_DIR_NAME + Path.SEPARATOR + this.jobContext.getJobId()); this.jobContext.getJobState().setJobLauncherType(LauncherTypeEnum.CLUSTER); this.stateSerDeRunnerThreads = Integer.parseInt(jobProps.getProperty(ParallelRunner.PARALLEL_RUNNER_THREADS_KEY, Integer.toString(ParallelRunner.DEFAULT_PARALLEL_RUNNER_THREADS))); Config stateStoreJobConfig = ConfigUtils.propertiesToConfig(jobProps) .withValue(ConfigurationKeys.STATE_STORE_FS_URI_KEY, ConfigValueFactory.fromAnyRef( new URI(appWorkDir.toUri().getScheme(), null, appWorkDir.toUri().getHost(), appWorkDir.toUri().getPort(), "/", null, null).toString())); this.stateStores = new StateStores(stateStoreJobConfig, appWorkDir, GobblinClusterConfigurationKeys.OUTPUT_TASK_STATE_DIR_NAME, appWorkDir, GobblinClusterConfigurationKeys.INPUT_WORK_UNIT_DIR_NAME, appWorkDir, GobblinClusterConfigurationKeys.JOB_STATE_DIR_NAME); URI fsUri = URI.create(jobProps.getProperty(ConfigurationKeys.FS_URI_KEY, ConfigurationKeys.LOCAL_FS_URI)); this.fs = FileSystem.get(fsUri, new Configuration()); this.taskStateCollectorService = new TaskStateCollectorService(jobProps, this.jobContext.getJobState(), this.eventBus, this.eventSubmitter, this.stateStores.getTaskStateStore(), this.outputTaskStateDir, this.getIssueRepository()); } @Override public void close() throws IOException { try { executeCancellation(); } finally { super.close(); } } public String getJobId() { return this.jobContext.getJobId(); } @Override protected void runWorkUnits(List<WorkUnit> workUnits) throws Exception { try { CountEventBuilder countEventBuilder = new CountEventBuilder(JobEvent.WORK_UNITS_CREATED, workUnits.size()); this.eventSubmitter.submit(countEventBuilder); log.info("Emitting WorkUnitsCreated Count: " + countEventBuilder.getCount()); long workUnitStartTime = System.currentTimeMillis(); workUnits.forEach((k) -> k.setProp(ConfigurationKeys.WORK_UNIT_CREATION_TIME_IN_MILLIS, workUnitStartTime)); // Start the output TaskState collector service this.taskStateCollectorService.startAsync().awaitRunning(); TimingEvent jobSubmissionTimer = this.eventSubmitter.getTimingEvent(TimingEvent.RunJobTimings.HELIX_JOB_SUBMISSION); synchronized (this.cancellationRequest) { if (!this.cancellationRequested) { submitJob(workUnits); jobSubmissionTimer.stop(); log.info(String.format("Submitted job %s", this.jobContext.getJobId())); this.jobSubmitted = true; } else { log.warn("Job {} not submitted as it was requested to be cancelled.", this.jobContext.getJobId()); } } TimingEvent jobRunTimer = this.eventSubmitter.getTimingEvent(TimingEvent.RunJobTimings.HELIX_JOB_RUN); waitJob(); jobRunTimer.stop(); log.info(String.format("Job %s completed", this.jobContext.getJobId())); } finally { // The last iteration of output TaskState collecting will run when the collector service gets stopped this.taskStateCollectorService.stopAsync().awaitTerminated(); cleanupWorkingDirectory(); } } protected void submitJob(List<WorkUnit> workUnits) throws Exception { } protected void waitJob() throws InterruptedException { } @Override protected void executeCancellation() { } public void launchJob(@Nullable JobListener jobListener) throws JobException { this.jobListener = jobListener; log.info("Launching Job"); boolean isLaunched = false; this.runningMap.putIfAbsent(this.jobContext.getJobName(), false); Throwable errorInJobLaunching = null; try { if (this.runningMap.replace(this.jobContext.getJobName(), false, true)) { log.info("Job {} will be executed, add into running map.", this.jobContext.getJobId()); isLaunched = true; launchJobImpl(jobListener); } else { log.warn("Job {} will not be executed because other jobs are still running.", this.jobContext.getJobId()); } // TODO: Better error handling. The current impl swallows exceptions for jobs that were started by this method call. // One potential way to improve the error handling is to make this error swallowing conifgurable } catch (Throwable t) { errorInJobLaunching = t; if (isLaunched) { // Attempts to cancel workflow if an error occurs during launch cancelJob(jobListener); } } finally { if (isLaunched) { if (this.runningMap.replace(this.jobContext.getJobName(), true, false)) { log.info("Job {} is done, remove from running map.", this.jobContext.getJobId()); } else { throw errorInJobLaunching == null ? new IllegalStateException( "A launched job should have running state equal to true in the running map.") : new RuntimeException("Failure in launching job:", errorInJobLaunching); } } } } /** * This method looks silly at first glance but exists for a reason. * * The method {@link GobblinJobLauncher#launchJob(JobListener)} contains boiler plate for handling exceptions and * mutating the runningMap to communicate state back to the {@link GobblinJobScheduler}. The boiler plate swallows * exceptions when launching the job because many use cases require that 1 job failure should not affect other jobs by causing the * entire process to fail through an uncaught exception. * * This method is useful for unit testing edge cases where we expect {@link JobException}s during the underlying launch operation. * It would be nice to not swallow exceptions, but the implications of doing that will require careful refactoring since * the class {@link GobblinJobLauncher} and {@link GobblinJobScheduler} are shared for 2 quite different cases * between GaaS and streaming. GaaS typically requiring many short lifetime workflows (where a failure is tolerated) and * streaming requiring a small number of long running workflows (where failure to submit is unexpected and is not * tolerated) * * @throws JobException */ @VisibleForTesting void launchJobImpl(@Nullable JobListener jobListener) throws JobException { super.launchJob(jobListener); } /** * Delete persisted {@link WorkUnit}s and {@link JobState} upon job completion. */ protected void cleanupWorkingDirectory() throws IOException { log.info("Deleting persisted work units for job " + this.jobContext.getJobId()); stateStores.getWuStateStore().delete(this.jobContext.getJobId()); // delete the directory that stores the task state files stateStores.getTaskStateStore().delete(outputTaskStateDir.getName()); log.info("Deleting job state file for job " + this.jobContext.getJobId()); if (this.stateStores.haveJobStateStore()) { this.stateStores.getJobStateStore().delete(this.jobContext.getJobId()); } else { Path jobStateFilePath = GobblinClusterUtils.getJobStateFilePath(false, this.appWorkDir, this.jobContext.getJobId()); this.fs.delete(jobStateFilePath, false); } } }
3,945
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting/work/SeqBackedWorkSpan.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.util.nesting.work; import java.util.Iterator; import java.util.List; import lombok.NoArgsConstructor; import lombok.NonNull; import lombok.RequiredArgsConstructor; /** Logical sub-sequence of `WORK_ITEM`s, backed for simplicity's sake by an in-memory collection */ @NoArgsConstructor @RequiredArgsConstructor public class SeqBackedWorkSpan<WORK_ITEM> implements Workload.WorkSpan<WORK_ITEM> { @NonNull private List<WORK_ITEM> elems; // CAUTION: despite the "warning: @NonNull is meaningless on a primitive @lombok.RequiredArgsConstructor"... // if removed, no two-arg ctor is generated, so syntax error on `new CollectionBackedTaskSpan(elems, startIndex)` @NonNull private int startingIndex; private transient Iterator<WORK_ITEM> statefulDelegatee = null; @Override public int getNumElems() { return elems.size(); } @Override public boolean hasNext() { if (statefulDelegatee == null) { statefulDelegatee = elems.iterator(); } return statefulDelegatee.hasNext(); } @Override public WORK_ITEM next() { if (statefulDelegatee == null) { throw new IllegalStateException("first call `hasNext()`!"); } return statefulDelegatee.next(); } @Override public String toString() { return getClassNickname() + "(" + startingIndex + "... {+" + getNumElems() + "})"; } protected String getClassNickname() { // return getClass().getSimpleName(); return "WorkSpan"; } }
3,946
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting/work/WorkflowAddr.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.util.nesting.work; import com.fasterxml.jackson.annotation.JsonIgnore; import com.google.common.base.Joiner; import com.google.common.collect.Lists; import java.util.ArrayList; import java.util.List; import lombok.Getter; import lombok.NoArgsConstructor; import lombok.NonNull; import lombok.RequiredArgsConstructor; /** Hierarchical address for nesting workflows (0-based). */ @NoArgsConstructor // IMPORTANT: for jackson (de)serialization @RequiredArgsConstructor public class WorkflowAddr { public static final String SEP = "."; /** initial, top-level address */ public static final WorkflowAddr ROOT = new WorkflowAddr(0); @Getter @NonNull // IMPORTANT: for jackson (de)serialization (which won't permit `final`) private List<Integer> segments; public WorkflowAddr(final int firstLevelOnly) { this(Lists.newArrayList(firstLevelOnly)); } /** @return 0-based depth */ @JsonIgnore // (because no-arg method resembles 'java bean property') public int getDepth() { return segments.size() - 1; } /** Create a child of the current `WFAddr` */ public WorkflowAddr createChild(int childLevel) { final List<Integer> copy = new ArrayList<>(segments); copy.add(childLevel); return new WorkflowAddr(copy); } @Override public String toString() { return Joiner.on(SEP).join(segments); } }
3,947
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting/work/Workload.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.util.nesting.work; import com.fasterxml.jackson.annotation.JsonTypeInfo; import java.util.Iterator; import java.util.Optional; /** * {@link Workload} models a logical collection of homogenous inputs over which a "foreach" operation can asynchronously apply * an arbitrary procedure to each element. This encapsulates "processing" the entire collection of sequential * "work item" specifications by the uniform application of the chosen procedure(s). * * Given Temporal's required determinism, the work items and work spans should remain unchanged, with stable sequential * ordering. This need not constrain `Workload`s to eager, advance elaboration: "streaming" definition is possible, * so long as producing a deterministic result. * * A actual, real-world workload might correspond to datastore contents, such as records serialized into HDFS files * or ordered DB query results. */ @JsonTypeInfo(use = JsonTypeInfo.Id.CLASS, include = JsonTypeInfo.As.PROPERTY, property = "@class") // to handle impls public interface Workload<WORK_ITEM> { /** * @return a sequential sub-sequence, from `startIndex` (0-based), unless it falls beyond the underlying sequence * NOTE: this is a blocking call that forces elaboration: `WorkSpan.getNumElems() < numElements` signifies end of seq */ Optional<WorkSpan<WORK_ITEM>> getSpan(int startIndex, int numElements); /** Non-blocking, best-effort advice: to support non-strict elaboration, does NOT guarantee `index` will not exceed */ boolean isIndexKnownToExceed(int index); default boolean isDefiniteSize() { return false; } /** Logical sub-sequence 'slice' of contiguous work items */ public interface WorkSpan<T> extends Iterator<T> { int getNumElems(); } }
3,948
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting/work/SeqSliceBackedWorkSpan.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.util.nesting.work; import java.util.NoSuchElementException; import lombok.NoArgsConstructor; import lombok.NonNull; import lombok.RequiredArgsConstructor; import com.fasterxml.jackson.annotation.JsonIgnore; /** Logical sub-sequence of `WORK_ITEM`s, backed for simplicity's sake by an in-memory collection, *SHARED* w/ other work spans */ @NoArgsConstructor @RequiredArgsConstructor public class SeqSliceBackedWorkSpan<WORK_ITEM> implements Workload.WorkSpan<WORK_ITEM> { private static final int NOT_SET_SENTINEL = -1; @NonNull private WORK_ITEM[] sharedElems; // CAUTION: despite the "warning: @NonNull is meaningless on a primitive @lombok.RequiredArgsConstructor"... // if removed, no two-arg ctor is generated, so syntax error on `new CollectionSliceBackedTaskSpan(elems, startIndex)` @NonNull private int startingIndex; @NonNull private int numElements; private transient volatile int nextElemIndex = NOT_SET_SENTINEL; @Override public int getNumElems() { return getEndingIndex() - startingIndex; } @Override public boolean hasNext() { if (nextElemIndex == NOT_SET_SENTINEL) { nextElemIndex = startingIndex; // NOTE: `startingIndex` should be effectively `final` (post-deser) and always >= 0 } return nextElemIndex < this.getEndingIndex(); } @Override public WORK_ITEM next() { if (nextElemIndex >= startingIndex + numElements) { throw new NoSuchElementException("index " + nextElemIndex + " >= " + startingIndex + " + " + numElements); } return sharedElems[nextElemIndex++]; } @Override public String toString() { return getClassNickname() + "(" + startingIndex + "... {+" + getNumElems() + "})"; } protected String getClassNickname() { // return getClass().getSimpleName(); return "WorkSpan"; } @JsonIgnore // (because no-arg method resembles 'java bean property') protected final int getEndingIndex() { return Math.min(startingIndex + numElements, sharedElems.length); } }
3,949
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting/workflow/NestingExecWorkflow.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.util.nesting.workflow; import java.util.Optional; import io.temporal.workflow.WorkflowInterface; import io.temporal.workflow.WorkflowMethod; import org.apache.gobblin.temporal.util.nesting.work.WorkflowAddr; import org.apache.gobblin.temporal.util.nesting.work.Workload; /** * Process all `WORK_ITEM`s of {@link Workload}, from `startIndex` to the end by creating child workflows, where this and * descendants should have at most `maxBranchesPerTree`, with at most `maxSubTreesPerTree` of those being child * workflows. (Non-child-workflow (terminal) branches are the activity executions.) * * The underlying motivation is to create logical workflows of unbounded size, despite Temporal's event history limit * of 50Ki events; see: https://docs.temporal.io/workflows#event-history * * IMPORTANT: `Math.sqrt(maxBranchesPerTree) == maxSubTreesPerTree` provides a good rule-of-thumb; `maxSubTreesPerTree * must not exceed that. This enables consolidation, wherein continued expansion occurs only along the tree's right-most edges. * * @param <WORK_ITEM> the type of task for which to invoke an appropriate activity * @param maxSubTreesForCurrentTreeOverride when the current tree should use different max sub-trees than descendants */ @WorkflowInterface public interface NestingExecWorkflow<WORK_ITEM> { /** @return the number of workload elements processed cumulatively by this Workflow and its children */ @WorkflowMethod int performWorkload( WorkflowAddr addr, Workload<WORK_ITEM> workload, int startIndex, int maxBranchesPerTree, int maxSubTreesPerTree, Optional<Integer> maxSubTreesForCurrentTreeOverride ); }
3,950
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/util/nesting/workflow/AbstractNestingExecWorkflowImpl.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.util.nesting.workflow; import java.time.Duration; import java.util.ArrayList; import java.util.Collections; import java.util.List; import java.util.Optional; import java.util.stream.Collectors; import java.util.stream.StreamSupport; import lombok.extern.slf4j.Slf4j; import org.apache.commons.compress.utils.Lists; import io.temporal.api.enums.v1.ParentClosePolicy; import io.temporal.workflow.Async; import io.temporal.workflow.ChildWorkflowOptions; import io.temporal.workflow.Promise; import io.temporal.workflow.Workflow; import org.apache.gobblin.temporal.util.nesting.work.WorkflowAddr; import org.apache.gobblin.temporal.util.nesting.work.Workload; /** Core skeleton of {@link NestingExecWorkflow}: realizing classes need only define {@link #launchAsyncActivity} */ @Slf4j public abstract class AbstractNestingExecWorkflowImpl<WORK_ITEM, ACTIVITY_RESULT> implements NestingExecWorkflow<WORK_ITEM> { public static final int NUM_SECONDS_TO_PAUSE_BEFORE_CREATING_SUB_TREE_DEFAULT = 10; public static final int MAX_CHILD_SUB_TREE_LEAVES_BEFORE_SHOULD_PAUSE_DEFAULT = 100; @Override public int performWorkload( final WorkflowAddr addr, final Workload<WORK_ITEM> workload, final int startIndex, final int maxBranchesPerTree, final int maxSubTreesPerTree, final Optional<Integer> maxSubTreesForCurrentTreeOverride ) { final int maxSubTreesForCurrent = maxSubTreesForCurrentTreeOverride.orElse(maxSubTreesPerTree); final int maxLeaves = maxBranchesPerTree - maxSubTreesForCurrent; final Optional<Workload.WorkSpan<WORK_ITEM>> optSpan = workload.getSpan(startIndex, maxLeaves); log.info("[" + addr + "] " + workload + " w/ start '" + startIndex + "'" + "; tree (" + maxBranchesPerTree + "/" + maxSubTreesPerTree + "): " + optSpan); if (!optSpan.isPresent()) { return 0; } else { final Workload.WorkSpan<WORK_ITEM> workSpan = optSpan.get(); final Iterable<WORK_ITEM> iterable = () -> workSpan; final List<Promise<ACTIVITY_RESULT>> childActivities = StreamSupport.stream(iterable.spliterator(), false) .map(t -> launchAsyncActivity(t)) .collect(Collectors.toList()); final List<Promise<Integer>> childSubTrees = new ArrayList<>(); if (workSpan.getNumElems() == maxLeaves) { // received as many as requested (did not stop short) int subTreeId = 0; for (int subTreeChildMaxSubTreesPerTree : consolidateSubTreeGrandChildren(maxSubTreesForCurrent, maxBranchesPerTree, maxSubTreesPerTree)) { // CAUTION: calc these *before* incrementing `subTreeId`! final int childStartIndex = startIndex + maxLeaves + (maxBranchesPerTree * subTreeId); final int nextChildId = maxLeaves + subTreeId; final WorkflowAddr childAddr = addr.createChild(nextChildId); final NestingExecWorkflow<WORK_ITEM> child = createChildWorkflow(childAddr); if (!workload.isIndexKnownToExceed(childStartIndex)) { // best-effort short-circuiting // IMPORTANT: insert pause before launch of each child workflow that may have direct leaves of its own. periodic pauses spread the load on the // temporal server, to avoid a sustained burst from submitting potentially very many async activities over the full hierarchical elaboration final int numDirectLeavesChildMayHave = maxBranchesPerTree - subTreeChildMaxSubTreesPerTree; if (numDirectLeavesChildMayHave > 0) { Workflow.sleep(calcPauseDurationBeforeCreatingSubTree(numDirectLeavesChildMayHave)); } childSubTrees.add( Async.function(child::performWorkload, childAddr, workload, childStartIndex, maxBranchesPerTree, maxSubTreesPerTree, Optional.of(subTreeChildMaxSubTreesPerTree))); ++subTreeId; } } } final Promise<Void> allActivityChildren = Promise.allOf(childActivities); allActivityChildren.get(); // ensure all complete prior to counting them in `overallActivitiesRollupCount` // TODO: determine whether any benefit to unordered `::get` blocking for any next ready (perhaps no difference...) final int descendantActivitiesRollupCount = childSubTrees.stream().map(Promise::get).reduce(0, (x, y) -> x + y); // TODO: consider a generalized reduce op for things other than counting! final int overallActivitiesRollupCount = workSpan.getNumElems() + descendantActivitiesRollupCount; log.info("[" + addr + "] activites finished coordinating: " + overallActivitiesRollupCount); return overallActivitiesRollupCount; } } /** Factory for invoking the specific activity by providing it args via {@link Async::function} */ protected abstract Promise<ACTIVITY_RESULT> launchAsyncActivity(WORK_ITEM task); protected NestingExecWorkflow<WORK_ITEM> createChildWorkflow(final WorkflowAddr childAddr) { // preserve the current workflow ID of this parent, but add the (hierarchical) address extension specific to each child String thisWorkflowId = Workflow.getInfo().getWorkflowId(); String childWorkflowId = thisWorkflowId.replaceAll("-[^-]+$", "") + "-" + childAddr; ChildWorkflowOptions childOpts = ChildWorkflowOptions.newBuilder() .setParentClosePolicy(ParentClosePolicy.PARENT_CLOSE_POLICY_ABANDON) .setWorkflowId(childWorkflowId) .build(); return Workflow.newChildWorkflowStub(NestingExecWorkflow.class, childOpts); } /** @return how long to pause prior to creating a child workflow, based on `numDirectLeavesChildMayHave` */ protected Duration calcPauseDurationBeforeCreatingSubTree(int numDirectLeavesChildMayHave) { // (only pause when an appreciable number of leaves) // TODO: use a configuration value, for simpler adjustment, rather than hard-code return numDirectLeavesChildMayHave > MAX_CHILD_SUB_TREE_LEAVES_BEFORE_SHOULD_PAUSE_DEFAULT ? Duration.ofSeconds(NUM_SECONDS_TO_PAUSE_BEFORE_CREATING_SUB_TREE_DEFAULT) : Duration.ZERO; } /** * "right-tilt" sub-tree's grandchildren, so final child gets all grandchildren (vs. constant grandchildren/child) * i.e. NOT!: * List<Integer> naiveUniformity = Collections.nCopies(numSubTreesPerSubTree, numSubTreeChildren); * @return each sub-tree's desired size, in ascending sub-tree order */ protected static List<Integer> consolidateSubTreeGrandChildren( final int numSubTreesPerSubTree, final int numChildrenTotal, final int numSubTreeChildren ) { if (numSubTreesPerSubTree <= 0) { return Lists.newArrayList(); } else if (isSqrt(numSubTreeChildren, numChildrenTotal)) { // redistribute all grandchild sub-trees to pack every grandchild beneath the final child sub-tree final List<Integer> grandChildCounts = new ArrayList<>(Collections.nCopies(numSubTreesPerSubTree - 1, 0)); grandChildCounts.add(numChildrenTotal); return grandChildCounts; } else { final int totalGrandChildSubTrees = numSubTreesPerSubTree * numSubTreeChildren; final int numTreesWithSolelySubTreeBranches = totalGrandChildSubTrees / numChildrenTotal; final int numSubTreesRemaining = totalGrandChildSubTrees % numChildrenTotal; assert (numTreesWithSolelySubTreeBranches == 1 && numSubTreesRemaining == 0) || numTreesWithSolelySubTreeBranches == 0 : "present limitation: at most one sub-tree may use further branching: (found: numSubTreesPerSubTree: " + numSubTreesPerSubTree + "; numChildrenTotal: " + numChildrenTotal + " / numSubTreeChildren: " + numSubTreeChildren + ")"; final List<Integer> grandChildCounts = new ArrayList<>(Collections.nCopies(numSubTreesPerSubTree - (numTreesWithSolelySubTreeBranches + 1), 0)); grandChildCounts.addAll(Collections.nCopies(Math.min(1, numSubTreesPerSubTree - numTreesWithSolelySubTreeBranches), numSubTreesRemaining)); grandChildCounts.addAll(Collections.nCopies(Math.min(numTreesWithSolelySubTreeBranches, numSubTreesPerSubTree), numChildrenTotal)); return grandChildCounts; } } /** @return whether `maxSubTrees` == `Math.sqrt(maxBranches)` */ private static boolean isSqrt(int maxSubTrees, int maxBranches) { return maxSubTrees > 0 && maxSubTrees * maxSubTrees == maxBranches; } }
3,951
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/launcher/ProcessWorkUnitsJobLauncher.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.launcher; import java.net.URI; import java.util.List; import java.util.Properties; import java.util.concurrent.ConcurrentHashMap; import lombok.extern.slf4j.Slf4j; import com.typesafe.config.ConfigFactory; import io.temporal.client.WorkflowOptions; import org.apache.hadoop.fs.Path; import org.apache.gobblin.metrics.Tag; import org.apache.gobblin.runtime.JobLauncher; import org.apache.gobblin.source.workunit.WorkUnit; import org.apache.gobblin.temporal.cluster.GobblinTemporalTaskRunner; import org.apache.gobblin.temporal.ddm.work.WUProcessingSpec; import org.apache.gobblin.temporal.ddm.work.assistance.Help; import org.apache.gobblin.temporal.ddm.workflow.ProcessWorkUnitsWorkflow; import org.apache.gobblin.temporal.joblauncher.GobblinTemporalJobLauncher; import org.apache.gobblin.temporal.joblauncher.GobblinTemporalJobScheduler; import org.apache.gobblin.util.PropertiesUtils; import static org.apache.gobblin.temporal.GobblinTemporalConfigurationKeys.GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_PREFIX; /** * A {@link JobLauncher} for the initial triggering of a Temporal workflow that executes {@link WorkUnit}s to fulfill * the work they specify. see: {@link ProcessWorkUnitsWorkflow} * * <p> * This class is instantiated by the {@link GobblinTemporalJobScheduler#buildJobLauncher(Properties)} on every job submission to launch the Gobblin job. * The actual task execution happens in the {@link GobblinTemporalTaskRunner}, usually in a different process. * </p> */ @Slf4j public class ProcessWorkUnitsJobLauncher extends GobblinTemporalJobLauncher { public static final String GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_NAME_NODE_URI = GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_PREFIX + "name.node.uri"; public static final String GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_WORK_UNITS_DIR = GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_PREFIX + "work.units.dir"; public static final String GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_WORK_MAX_BRANCHES_PER_TREE = GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_PREFIX + "work.max.branches.per.tree"; public static final String GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_WORK_MAX_SUB_TREES_PER_TREE = GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_PREFIX + "work.max.sub.trees.per.tree"; public static final String WORKFLOW_ID_BASE = "ProcessWorkUnits"; public ProcessWorkUnitsJobLauncher( Properties jobProps, Path appWorkDir, List<? extends Tag<?>> metadataTags, ConcurrentHashMap<String, Boolean> runningMap ) throws Exception { super(jobProps, appWorkDir, metadataTags, runningMap); } @Override public void submitJob(List<WorkUnit> workunits) { try { URI nameNodeUri = new URI(PropertiesUtils.getRequiredProp(this.jobProps, GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_NAME_NODE_URI)); // NOTE: `Path` is challenging for temporal to ser/de, but nonetheless do pre-construct as `Path`, to pre-validate this prop string's contents Path workUnitsDir = new Path(PropertiesUtils.getRequiredProp(this.jobProps, GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_WORK_UNITS_DIR)); WUProcessingSpec wuSpec = new WUProcessingSpec(nameNodeUri, workUnitsDir.toString()); if (this.jobProps.containsKey(GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_WORK_MAX_BRANCHES_PER_TREE) && this.jobProps.containsKey(GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_WORK_MAX_SUB_TREES_PER_TREE)) { int maxBranchesPerTree = PropertiesUtils.getRequiredPropAsInt(this.jobProps, GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_WORK_MAX_BRANCHES_PER_TREE); int maxSubTreesPerTree = PropertiesUtils.getRequiredPropAsInt(this.jobProps, GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_WORK_MAX_SUB_TREES_PER_TREE); wuSpec.setTuning(new WUProcessingSpec.Tuning(maxBranchesPerTree, maxSubTreesPerTree)); } WorkflowOptions options = WorkflowOptions.newBuilder() .setTaskQueue(this.queueName) .setWorkflowId(Help.qualifyNamePerExec(WORKFLOW_ID_BASE, wuSpec, ConfigFactory.parseProperties(jobProps))) .build(); ProcessWorkUnitsWorkflow workflow = this.client.newWorkflowStub(ProcessWorkUnitsWorkflow.class, options); workflow.process(wuSpec); } catch (Exception e) { throw new RuntimeException(e); } } }
3,952
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/util/JobStateUtils.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.util; import java.util.Properties; import java.util.concurrent.ExecutionException; import lombok.extern.slf4j.Slf4j; import com.google.common.cache.Cache; import com.google.common.cache.CacheBuilder; import com.typesafe.config.ConfigFactory; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.apache.gobblin.broker.SharedResourcesBrokerFactory; import org.apache.gobblin.broker.gobblin_scopes.GobblinScopeTypes; import org.apache.gobblin.broker.gobblin_scopes.JobScopeInstance; import org.apache.gobblin.broker.iface.SharedResourcesBroker; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.metastore.FsStateStore; import org.apache.gobblin.metastore.StateStore; import org.apache.gobblin.runtime.JobState; import org.apache.gobblin.runtime.TaskState; /** * Utilities for applying {@link JobState} info to various ends: * - creating a {@link SharedResourcesBroker} * - obtaining a {@link StateStore<TaskState>} */ @Slf4j public class JobStateUtils { private static final String OUTPUT_DIR_NAME = "output"; // following MRJobLauncher.OUTPUT_DIR_NAME // reuse same handle among activities executed by the same worker private static final transient Cache<Path, StateStore<TaskState>> taskStateStoreByPath = CacheBuilder.newBuilder().build(); private JobStateUtils() {} public static StateStore<TaskState> openTaskStateStore(JobState jobState, FileSystem fs) { try { Path taskStateStorePath = JobStateUtils.getTaskStateStorePath(jobState, fs); return taskStateStoreByPath.get(taskStateStorePath, () -> openTaskStateStoreUncached(jobState, fs) ); } catch (ExecutionException ee) { throw new RuntimeException(ee); } } public static StateStore<TaskState> openTaskStateStoreUncached(JobState jobState, FileSystem fs) { Path taskStateStorePath = JobStateUtils.getTaskStateStorePath(jobState, fs); log.info("opening FS task state store at path '{}'", taskStateStorePath); return new FsStateStore<>(fs, taskStateStorePath.toUri().getPath(), TaskState.class); } /** * ATTENTION: derives path according to {@link org.apache.gobblin.runtime.mapreduce.MRJobLauncher} conventions, using same * {@link ConfigurationKeys#MR_JOB_ROOT_DIR_KEY} * @return path to {@link FsStateStore<TaskState>} backing dir */ public static Path getTaskStateStorePath(JobState jobState, FileSystem fs) { Properties jobProps = jobState.getProperties(); Path jobOutputPath = new Path( new Path( new Path( jobProps.getProperty(ConfigurationKeys.MR_JOB_ROOT_DIR_KEY), JobState.getJobNameFromProps(jobProps)), JobState.getJobIdFromProps(jobProps)), OUTPUT_DIR_NAME); return fs.makeQualified(jobOutputPath); } public static SharedResourcesBroker<GobblinScopeTypes> getSharedResourcesBroker(JobState jobState) { SharedResourcesBroker<GobblinScopeTypes> globalBroker = SharedResourcesBrokerFactory.createDefaultTopLevelBroker( ConfigFactory.parseProperties(jobState.getProperties()), GobblinScopeTypes.GLOBAL.defaultScopeInstance()); return globalBroker.newSubscopedBuilder(new JobScopeInstance(jobState.getJobName(), jobState.getJobId())).build(); } }
3,953
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/activity/ProcessWorkUnit.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.activity; import io.temporal.activity.ActivityInterface; import io.temporal.activity.ActivityMethod; import org.apache.gobblin.temporal.ddm.work.WorkUnitClaimCheck; /** Activity for processing/executing a {@link org.apache.gobblin.source.workunit.WorkUnit}, provided by claim-check */ @ActivityInterface public interface ProcessWorkUnit { @ActivityMethod // CAUTION: void return type won't work, as apparently it mayn't be the return type for `io.temporal.workflow.Functions.Func1`! int processWorkUnit(WorkUnitClaimCheck wu); }
3,954
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/activity
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/activity/impl/ProcessWorkUnitImpl.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.activity.impl; import java.io.IOException; import java.util.List; import java.util.Optional; import java.util.Properties; import java.util.function.Predicate; import java.util.stream.Collectors; import lombok.extern.slf4j.Slf4j; import com.google.common.collect.Lists; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.apache.gobblin.broker.gobblin_scopes.GobblinScopeTypes; import org.apache.gobblin.broker.iface.SharedResourcesBroker; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.configuration.State; import org.apache.gobblin.configuration.WorkUnitState; import org.apache.gobblin.data.management.copy.CopyEntity; import org.apache.gobblin.data.management.copy.CopySource; import org.apache.gobblin.data.management.copy.CopyableFile; import org.apache.gobblin.metastore.StateStore; import org.apache.gobblin.runtime.AbstractTaskStateTracker; import org.apache.gobblin.runtime.GobblinMultiTaskAttempt; import org.apache.gobblin.runtime.JobState; import org.apache.gobblin.runtime.Task; import org.apache.gobblin.runtime.TaskCreationException; import org.apache.gobblin.runtime.TaskExecutor; import org.apache.gobblin.runtime.TaskState; import org.apache.gobblin.runtime.TaskStateTracker; import org.apache.gobblin.runtime.troubleshooter.AutomaticTroubleshooter; import org.apache.gobblin.runtime.troubleshooter.NoopAutomaticTroubleshooter; import org.apache.gobblin.source.workunit.WorkUnit; import org.apache.gobblin.temporal.ddm.activity.ProcessWorkUnit; import org.apache.gobblin.temporal.ddm.util.JobStateUtils; import org.apache.gobblin.temporal.ddm.work.assistance.Help; import org.apache.gobblin.temporal.ddm.work.WorkUnitClaimCheck; import org.apache.gobblin.util.JobLauncherUtils; @Slf4j public class ProcessWorkUnitImpl implements ProcessWorkUnit { private static final int LOG_EXTENDED_PROPS_EVERY_WORK_UNITS_STRIDE = 100; @Override public int processWorkUnit(WorkUnitClaimCheck wu) { try (FileSystem fs = Help.loadFileSystemForce(wu)) { List<WorkUnit> workUnits = loadFlattenedWorkUnits(wu, fs); log.info("WU [{}] - loaded {} workUnits", wu.getCorrelator(), workUnits.size()); JobState jobState = Help.loadJobState(wu, fs); return execute(workUnits, wu, jobState, fs); } catch (IOException | InterruptedException e) { throw new RuntimeException(e); } } protected List<WorkUnit> loadFlattenedWorkUnits(WorkUnitClaimCheck wu, FileSystem fs) throws IOException { Path wuPath = new Path(wu.getWorkUnitPath()); WorkUnit workUnit = JobLauncherUtils.createEmptyWorkUnitPerExtension(wuPath); Help.deserializeStateWithRetries(fs, wuPath, workUnit, wu); return JobLauncherUtils.flattenWorkUnits(Lists.newArrayList(workUnit)); } /** * NOTE: adapted from {@link org.apache.gobblin.runtime.mapreduce.MRJobLauncher.TaskRunner#run(org.apache.hadoop.mapreduce.Mapper.Context)} * @return count of how many tasks executed (0 if execution ultimately failed, but we *believe* TaskState should already have been recorded beforehand) */ protected int execute(List<WorkUnit> workUnits, WorkUnitClaimCheck wu, JobState jobState, FileSystem fs) throws IOException, InterruptedException { String containerId = "container-id-for-wu-" + wu.getCorrelator(); StateStore<TaskState> taskStateStore = Help.openTaskStateStore(wu, fs); TaskStateTracker taskStateTracker = createEssentializedTaskStateTracker(wu); TaskExecutor taskExecutor = new TaskExecutor(new Properties()); GobblinMultiTaskAttempt.CommitPolicy multiTaskAttemptCommitPolicy = GobblinMultiTaskAttempt.CommitPolicy.IMMEDIATE; // as no speculative exec SharedResourcesBroker<GobblinScopeTypes> resourcesBroker = JobStateUtils.getSharedResourcesBroker(jobState); AutomaticTroubleshooter troubleshooter = new NoopAutomaticTroubleshooter(); // AutomaticTroubleshooterFactory.createForJob(ConfigUtils.propertiesToConfig(wu.getStateConfig().getProperties())); troubleshooter.start(); List<String> fileSourcePaths = workUnits.stream() .map(workUnit -> describeAsCopyableFile(workUnit, wu.getWorkUnitPath())) .collect(Collectors.toList()); log.info("WU [{}] - submitting {} workUnits for copying files: {}", wu.getCorrelator(), workUnits.size(), fileSourcePaths); log.debug("WU [{}] - (first) workUnit: {}", wu.getCorrelator(), workUnits.get(0).toJsonString()); try { GobblinMultiTaskAttempt taskAttempt = GobblinMultiTaskAttempt.runWorkUnits( jobState.getJobId(), containerId, jobState, workUnits, taskStateTracker, taskExecutor, taskStateStore, multiTaskAttemptCommitPolicy, resourcesBroker, troubleshooter.getIssueRepository(), createInterruptionPredicate(fs, jobState)); return taskAttempt.getNumTasksCreated(); } catch (TaskCreationException tce) { // derived type of `IOException` that ought not be caught! throw tce; } catch (IOException ioe) { // presume execution already occurred, with `TaskState` written to reflect outcome log.warn("WU [" + wu.getCorrelator() + "] - continuing on despite IOException:", ioe); return 0; } } /** Demonstration processing, to isolate debugging of WU loading and deserialization */ protected int countSumProperties(List<WorkUnit> workUnits, WorkUnitClaimCheck wu) { int totalNumProps = workUnits.stream().mapToInt(workUnit -> workUnit.getPropertyNames().size()).sum(); log.info("opened WU [{}] to find {} properties total at '{}'", wu.getCorrelator(), totalNumProps, wu.getWorkUnitPath()); return totalNumProps; } protected TaskStateTracker createEssentializedTaskStateTracker(WorkUnitClaimCheck wu) { return new AbstractTaskStateTracker(new Properties(), log) { @Override public void registerNewTask(Task task) { // TODO: shall we schedule metrics update based on config? } @Override public void onTaskRunCompletion(Task task) { task.markTaskCompletion(); } @Override public void onTaskCommitCompletion(Task task) { TaskState taskState = task.getTaskState(); // TODO: if metrics configured, report them now log.info("WU [{} = {}] - finished commit after {}ms with state {}{}", wu.getCorrelator(), task.getTaskId(), taskState.getTaskDuration(), taskState.getWorkingState(), taskState.getWorkingState().equals(WorkUnitState.WorkingState.SUCCESSFUL) ? (" to: " + taskState.getProp(ConfigurationKeys.WRITER_OUTPUT_DIR)) : ""); log.debug("WU [{} = {}] - task state: {}", wu.getCorrelator(), task.getTaskId(), taskState.toJsonString(shouldUseExtendedLogging(wu))); getOptCopyableFile(taskState).ifPresent(copyableFile -> { log.info("WU [{} = {}] - completed copyableFile: {}", wu.getCorrelator(), task.getTaskId(), copyableFile.toJsonString(shouldUseExtendedLogging(wu))); }); } }; } protected String describeAsCopyableFile(WorkUnit workUnit, String workUnitPath) { return getOptFirstCopyableFile(Lists.newArrayList(workUnit), workUnitPath) .map(copyableFile -> copyableFile.getOrigin().getPath().toString()) .orElse( "<<not a CopyableFile(" + getOptCopyEntityClass(workUnit, workUnitPath) .map(Class::getSimpleName) .orElse("<<not a CopyEntity!>>") + "): '" + workUnitPath + "'" ); } protected Optional<CopyableFile> getOptCopyableFile(TaskState taskState) { return getOptCopyableFile(taskState, "taskState '" + taskState.getTaskId() + "'"); } protected Optional<CopyableFile> getOptFirstCopyableFile(List<WorkUnit> workUnits, String workUnitPath) { return Optional.of(workUnits).filter(wus -> wus.size() > 0).flatMap(wus -> getOptCopyableFile(wus.get(0), "workUnit '" + workUnitPath + "'") ); } protected Optional<CopyableFile> getOptCopyableFile(State state, String logDesc) { return getOptCopyEntityClass(state, logDesc).flatMap(copyEntityClass -> { log.debug("(state) {} got (copyEntity) {}", state.getClass().getName(), copyEntityClass.getName()); if (CopyableFile.class.isAssignableFrom(copyEntityClass)) { String serialization = state.getProp(CopySource.SERIALIZED_COPYABLE_FILE); if (serialization != null) { return Optional.of((CopyableFile) CopyEntity.deserialize(serialization)); } } return Optional.empty(); }); } protected Optional<Class<?>> getOptCopyEntityClass(State state, String logDesc) { try { return Optional.of(CopySource.getCopyEntityClass(state)); } catch (IOException ioe) { log.warn(logDesc + " - failed getting copy entity class:", ioe); return Optional.empty(); } } protected Predicate<GobblinMultiTaskAttempt> createInterruptionPredicate(FileSystem fs, JobState jobState) { // TODO - decide whether to support... and if so, employ a useful path; otherwise, just evaluate predicate to always false Path interruptionPath = new Path("/not/a/real/path/that/should/ever/exist!"); return createInterruptionPredicate(fs, interruptionPath); } protected Predicate<GobblinMultiTaskAttempt> createInterruptionPredicate(FileSystem fs, Path interruptionPath) { return (gmta) -> { try { return fs.exists(interruptionPath); } catch (IOException ioe) { return false; } }; } protected boolean shouldUseExtendedLogging(WorkUnitClaimCheck wu) { try { return Long.parseLong(wu.getCorrelator()) % LOG_EXTENDED_PROPS_EVERY_WORK_UNITS_STRIDE == 0; } catch (NumberFormatException nfe) { log.warn("unexpected, non-numeric correlator: '{}'", wu.getCorrelator()); return false; } } }
3,955
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work/EagerFsDirBackedWorkUnitClaimCheckWorkload.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.work; import java.net.URI; import java.util.Comparator; import com.fasterxml.jackson.annotation.JsonIgnore; import org.apache.hadoop.fs.FileStatus; /** * {@link AbstractEagerFsDirBackedWorkload} for {@link WorkUnitClaimCheck} `WORK_ITEM`s, which uses {@link WorkUnitClaimCheck#getWorkUnitPath()} * for their total-ordering. */ @lombok.NoArgsConstructor // IMPORTANT: for jackson (de)serialization @lombok.ToString(callSuper = true) public class EagerFsDirBackedWorkUnitClaimCheckWorkload extends AbstractEagerFsDirBackedWorkload<WorkUnitClaimCheck> { public EagerFsDirBackedWorkUnitClaimCheckWorkload(URI fileSystemUri, String hdfsDir) { super(fileSystemUri, hdfsDir); } @Override protected WorkUnitClaimCheck fromFileStatus(FileStatus fileStatus) { // begin by setting all correlators to empty return new WorkUnitClaimCheck("", this.getFileSystemUri(), fileStatus.getPath().toString()); } @Override @JsonIgnore // (because no-arg method resembles 'java bean property') protected Comparator<WorkUnitClaimCheck> getWorkItemComparator() { return Comparator.comparing(WorkUnitClaimCheck::getWorkUnitPath); } @Override protected void acknowledgeOrdering(int index, WorkUnitClaimCheck item) { // later, after the post-total-ordering indices are know, use each item's index as its correlator item.setCorrelator(Integer.toString(index)); } }
3,956
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work/AbstractEagerFsDirBackedWorkload.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.work; import java.io.FileNotFoundException; import java.io.IOException; import java.net.URI; import java.util.Arrays; import java.util.Comparator; import java.util.Optional; import java.util.stream.IntStream; import java.util.stream.Stream; import lombok.AccessLevel; import lombok.Getter; import lombok.NonNull; import lombok.Setter; import lombok.extern.slf4j.Slf4j; import com.fasterxml.jackson.annotation.JsonIgnore; import org.apache.hadoop.fs.FileStatus; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.apache.hadoop.fs.PathFilter; import org.apache.gobblin.configuration.State; import org.apache.gobblin.temporal.ddm.work.styles.FileSystemApt; import org.apache.gobblin.temporal.util.nesting.work.SeqSliceBackedWorkSpan; import org.apache.gobblin.temporal.util.nesting.work.Workload; import org.apache.gobblin.util.HadoopUtils; /** * {@link Workload} of `WORK_ITEM`s (as defined by derived class) that originates from the eagerly loaded contents of * the directory `fsDir` within the {@link FileSystem} at `nameNodeUri`. * * IMPORTANT: to abide by Temporal's required determinism, a derived class must provide a {@link Comparator} for the * *total ordering* of `WORK_ITEM`s. */ @lombok.NoArgsConstructor // IMPORTANT: for jackson (de)serialization @lombok.RequiredArgsConstructor @lombok.ToString(exclude = { "stateConfig", "cachedWorkItems" }) @Slf4j public abstract class AbstractEagerFsDirBackedWorkload<WORK_ITEM> implements Workload<WORK_ITEM>, FileSystemApt { @Getter @NonNull private URI fileSystemUri; // NOTE: use `String` rather than `Path` to avoid: com.fasterxml.jackson.databind.exc.MismatchedInputException: // Cannot construct instance of `org.apache.hadoop.fs.Path` (although at least one Creator exists): // cannot deserialize from Object value (no delegate- or property-based Creator) @NonNull private String fsDir; @Getter(AccessLevel.PROTECTED) @Setter(AccessLevel.PROTECTED) private transient volatile WORK_ITEM[] cachedWorkItems = null; @Override public Optional<Workload.WorkSpan<WORK_ITEM>> getSpan(final int startIndex, final int numElements) { WORK_ITEM[] workItems = getCachedWorkItems(); if (startIndex >= workItems.length || startIndex < 0) { return Optional.empty(); } else { return Optional.of(new SeqSliceBackedWorkSpan<>(workItems, startIndex, numElements)); } } @Override public boolean isIndexKnownToExceed(final int index) { return isDefiniteSize() && cachedWorkItems != null && index >= cachedWorkItems.length; } @Override @JsonIgnore // (because no-arg method resembles 'java bean property') public boolean isDefiniteSize() { return true; } protected abstract WORK_ITEM fromFileStatus(FileStatus fileStatus); /** * IMPORTANT: to satisfy Temporal's required determinism, the `WORK_ITEM`s need a consistent total ordering * WARNING: this works so long as dir contents are unchanged in iterim * TODO: handle case of dir contents growing (e.g. use timestamp to filter out newer paths)... how could we handle the case of shrinking/deletion? */ @JsonIgnore // (because no-arg method resembles 'java bean property') protected abstract Comparator<WORK_ITEM> getWorkItemComparator(); /** Hook for each `WORK_ITEM` to be associated with its final, post-sorting ordinal index */ protected void acknowledgeOrdering(int index, WORK_ITEM workItem) { // no-op } @JsonIgnore // (because no-arg method resembles 'java bean property') protected PathFilter getPathFilter() { return f -> true; } @JsonIgnore // (because no-arg method resembles 'java bean property') protected final synchronized WORK_ITEM[] getCachedWorkItems() { if (cachedWorkItems != null) { return cachedWorkItems; } try (FileSystem fs = loadFileSystem()) { FileStatus[] fileStatuses = fs.listStatus(new Path(fsDir), this.getPathFilter()); log.info("loaded {} paths from '{}'", fileStatuses.length, fsDir); WORK_ITEM[] workItems = (WORK_ITEM[])Stream.of(fileStatuses).map(this::fromFileStatus).toArray(Object[]::new); sortWorkItems(workItems); IntStream.range(0, workItems.length) .forEach(i -> this.acknowledgeOrdering(i, workItems[i])); cachedWorkItems = workItems; return cachedWorkItems; } catch (FileNotFoundException fnfe) { throw new RuntimeException("directory not found: '" + fsDir + "'"); } catch (IOException ioe) { throw new RuntimeException(ioe); } } @JsonIgnore // (because no-arg method resembles 'java bean property') @Override public State getFileSystemConfig() { return new State(); // TODO - figure out how to truly set! } @JsonIgnore // (because no-arg method resembles 'java bean property') protected FileSystem loadFileSystem() throws IOException { return HadoopUtils.getFileSystem(this.fileSystemUri, this.getFileSystemConfig()); } private void sortWorkItems(WORK_ITEM[] workItems) { Arrays.sort(workItems, getWorkItemComparator()); } }
3,957
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work/WUProcessingSpec.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.work; import java.net.URI; import java.util.Optional; import lombok.Data; import lombok.NoArgsConstructor; import lombok.NonNull; import lombok.RequiredArgsConstructor; import com.fasterxml.jackson.annotation.JsonIgnore; import org.apache.hadoop.fs.Path; import org.apache.gobblin.configuration.State; import org.apache.gobblin.runtime.AbstractJobLauncher; import org.apache.gobblin.temporal.ddm.work.styles.FileSystemApt; import org.apache.gobblin.temporal.ddm.work.styles.FileSystemJobStateful; import org.apache.gobblin.temporal.util.nesting.work.WorkflowAddr; import org.apache.gobblin.temporal.util.nesting.work.Workload; /** * Intended to reference multiple {@link org.apache.gobblin.source.workunit.WorkUnit}s to process, where `workUnitsDir` * is resolved against the {@link org.apache.hadoop.fs.FileSystem} given by `nameNodeUri`. see: */ @Data @NoArgsConstructor // IMPORTANT: for jackson (de)serialization @RequiredArgsConstructor public class WUProcessingSpec implements FileSystemApt, FileSystemJobStateful { @NonNull private URI fileSystemUri; @NonNull private String workUnitsDir; @NonNull private Tuning tuning = Tuning.DEFAULT; @JsonIgnore // (because no-arg method resembles 'java bean property') @Override public State getFileSystemConfig() { return new State(); // TODO - figure out how to truly set! } @JsonIgnore // (because no-arg method resembles 'java bean property') @Override public Path getJobStatePath() { // TODO: decide whether wise to hard-code... (per `MRJobLauncher` conventions, we expect job state file to be sibling of WU dir) return new Path(new Path(workUnitsDir).getParent(), AbstractJobLauncher.JOB_STATE_FILE_NAME); } /** Configuration for {@link org.apache.gobblin.temporal.util.nesting.workflow.NestingExecWorkflow#performWorkload(WorkflowAddr, Workload, int, int, int, Optional)}*/ @Data @NoArgsConstructor // IMPORTANT: for jackson (de)serialization @RequiredArgsConstructor public static class Tuning { public static final int DEFAULT_MAX_BRANCHES_PER_TREE = 900; public static final int DEFAULT_SUB_TREES_PER_TREE = 30; public static final Tuning DEFAULT = new Tuning(DEFAULT_MAX_BRANCHES_PER_TREE, DEFAULT_SUB_TREES_PER_TREE); @NonNull private int maxBranchesPerTree; @NonNull private int maxSubTreesPerTree; } }
3,958
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work/WorkUnitClaimCheck.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.work; import java.net.URI; import lombok.Data; import lombok.NoArgsConstructor; import lombok.NonNull; import lombok.RequiredArgsConstructor; import com.fasterxml.jackson.annotation.JsonIgnore; import org.apache.hadoop.fs.Path; import org.apache.gobblin.configuration.State; import org.apache.gobblin.runtime.AbstractJobLauncher; import org.apache.gobblin.temporal.ddm.work.styles.FileSystemApt; import org.apache.gobblin.temporal.ddm.work.styles.FileSystemJobStateful; /** * Conveys a {@link org.apache.gobblin.source.workunit.WorkUnit} by claim-check, where the `workUnitPath` is resolved * against the {@link org.apache.hadoop.fs.FileSystem} given by `nameNodeUri`. see: * @see <a href="https://learn.microsoft.com/en-us/azure/architecture/patterns/claim-check">Claim-Check Pattern</a> */ @Data @NoArgsConstructor // IMPORTANT: for jackson (de)serialization @RequiredArgsConstructor public class WorkUnitClaimCheck implements FileSystemApt, FileSystemJobStateful { @NonNull private String correlator; @NonNull private URI fileSystemUri; @NonNull private String workUnitPath; @JsonIgnore // (because no-arg method resembles 'java bean property') @Override public State getFileSystemConfig() { return new State(); // TODO - figure out how to truly set! } @JsonIgnore // (because no-arg method resembles 'java bean property') @Override public Path getJobStatePath() { // TODO: decide whether wise to hard-code... (per `MRJobLauncher` conventions, we expect job state file to be sibling of WU dir) return new Path(new Path(workUnitPath).getParent().getParent(), AbstractJobLauncher.JOB_STATE_FILE_NAME); } }
3,959
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work/assistance/Help.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.work.assistance; import java.io.IOException; import java.net.URI; import java.util.Optional; import java.util.UUID; import java.util.concurrent.atomic.AtomicInteger; import java.util.concurrent.ExecutionException; import lombok.extern.slf4j.Slf4j; import com.google.common.cache.Cache; import com.google.common.cache.CacheBuilder; import com.typesafe.config.Config; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.configuration.State; import org.apache.gobblin.metastore.StateStore; import org.apache.gobblin.runtime.JobState; import org.apache.gobblin.runtime.TaskState; import org.apache.gobblin.temporal.ddm.util.JobStateUtils; import org.apache.gobblin.temporal.ddm.work.styles.FileSystemJobStateful; import org.apache.gobblin.temporal.ddm.work.styles.FileSystemApt; import org.apache.gobblin.temporal.ddm.work.styles.JobStateful; import org.apache.gobblin.util.HadoopUtils; import org.apache.gobblin.util.SerializationUtils; /** Various capabilities useful in implementing Distributed Data Movement (DDM) */ @Slf4j public class Help { public static final int MAX_DESERIALIZATION_FS_LOAD_ATTEMPTS = 5; public static final int LOG_CACHE_STATS_EVERY_N_ACCESSES = 1000; public static final String AZKABAN_FLOW_EXEC_ID_KEY = "azkaban.flow.execid"; public static final String USER_TO_PROXY_KEY = "user.to.proxy"; // treat `JobState` as immutable and cache, for reuse among activities executed by the same worker private static final transient Cache<Path, JobState> jobStateByPath = CacheBuilder.newBuilder().recordStats().build(); private static final transient AtomicInteger jobStateAccessCount = new AtomicInteger(0); private Help() {} public static String qualifyNamePerExec(String name, FileSystemJobStateful f, Config workerConfig) { return name + "_" + calcPerExecQualifier(f, workerConfig); } public static String qualifyNamePerExec(String name, Config workerConfig) { return name + "_" + calcPerExecQualifier(workerConfig); } public static String calcPerExecQualifier(FileSystemJobStateful f, Config workerConfig) { Optional<String> optFlowExecId = Optional.empty(); try { optFlowExecId = Optional.of(loadJobState(f).getProp(ConfigurationKeys.FLOW_EXECUTION_ID_KEY, null)); } catch (IOException e) { log.warn("unable to loadJobState", e); } return optFlowExecId.map(x -> x + "_").orElse("") + calcPerExecQualifier(workerConfig); } public static String calcPerExecQualifier(Config workerConfig) { String userToProxy = workerConfig.hasPath(USER_TO_PROXY_KEY) ? workerConfig.getString(USER_TO_PROXY_KEY) : ""; String azFlowExecId = workerConfig.hasPath(AZKABAN_FLOW_EXEC_ID_KEY) ? workerConfig.getString(AZKABAN_FLOW_EXEC_ID_KEY) : UUID.randomUUID().toString(); return userToProxy + "_" + azFlowExecId; } public static FileSystem loadFileSystem(FileSystemApt a) throws IOException { // NOTE: `FileSystem.get` appears to implement caching, which should facilitate sharing among activities executing on the same worker return loadFileSystemForUri(a.getFileSystemUri(), a.getFileSystemConfig()); } public static FileSystem loadFileSystemForUri(URI fsUri, State fsConfig) throws IOException { // TODO - determine whether this works... unclear whether it led to "FS closed", or that had another cause... // return HadoopUtils.getFileSystem(fsUri, fsConfig); Configuration conf = HadoopUtils.getConfFromState(fsConfig); return FileSystem.get(fsUri, conf); } public static FileSystem loadFileSystemForce(FileSystemApt a) throws IOException { return loadFileSystemForUriForce(a.getFileSystemUri(), a.getFileSystemConfig()); } public static FileSystem loadFileSystemForUriForce(URI fsUri, State fsConfig) throws IOException { // for reasons still not fully understood, we encountered many "FS closed" failures before disabling HDFS caching--especially as num WUs increased. // perhaps caching-facilitated reuse of the same FS across multiple WUs caused prior WU execs to leave the FS in a problematic state for subsequent execs // TODO - more investigation to sort out the true RC... and whether caching definitively is or is not possible for use here! // return HadoopUtils.getFileSystem(fsUri, fsConfig); Configuration conf = HadoopUtils.getConfFromState(fsConfig); conf.setBoolean("fs.hdfs.impl.disable.cache", true); return FileSystem.get(fsUri, conf); } public static JobState loadJobState(FileSystemJobStateful f) throws IOException { try (FileSystem fs = loadFileSystemForce(f)) { return loadJobState(f, fs); } } public static JobState loadJobState(JobStateful js, FileSystem fs) throws IOException { try { incrementJobStateAccess(); return jobStateByPath.get(js.getJobStatePath(), () -> loadJobStateUncached(js, fs) ); } catch (ExecutionException ee) { throw new IOException(ee); } } public static JobState loadJobStateUncached(JobStateful js, FileSystem fs) throws IOException { JobState jobState = new JobState(); SerializationUtils.deserializeState(fs, js.getJobStatePath(), jobState); log.info("loaded jobState from '{}': {}", js.getJobStatePath(), jobState.toJsonString(true)); return jobState; } public static JobState loadJobStateWithRetries(FileSystemJobStateful f) throws IOException { try (FileSystem fs = loadFileSystemForce(f)) { return loadJobStateWithRetries(f, fs); } } public static JobState loadJobStateWithRetries(FileSystemJobStateful f, FileSystem fs) throws IOException { try { incrementJobStateAccess(); return jobStateByPath.get(f.getJobStatePath(), () -> loadJobStateUncachedWithRetries(f, fs, f) ); } catch (ExecutionException ee) { throw new IOException(ee); } } public static JobState loadJobStateUncachedWithRetries(JobStateful js, FileSystem fs, FileSystemApt fsApt) throws IOException { JobState jobState = new JobState(); deserializeStateWithRetries(fs, js.getJobStatePath(), jobState, fsApt, MAX_DESERIALIZATION_FS_LOAD_ATTEMPTS); log.info("loaded jobState from '{}': {}", js.getJobStatePath(), jobState.toJsonString(true)); return jobState; } public static <T extends State> void deserializeStateWithRetries(FileSystem fs, Path path, T state, FileSystemApt fsApt) throws IOException { deserializeStateWithRetries(fs, path, state, fsApt, MAX_DESERIALIZATION_FS_LOAD_ATTEMPTS); } // TODO: decide whether actually necessary... it was added in a fit of debugging "FS closed" errors public static <T extends State> void deserializeStateWithRetries(FileSystem fs, Path path, T state, FileSystemApt fsApt, int maxAttempts) throws IOException { for (int i = 0; i < maxAttempts; ++i) { if (i > 0) { log.info("reopening FS '{}' to retry ({}) deserialization (attempt {})", fsApt.getFileSystemUri(), state.getClass().getSimpleName(), i); fs = Help.loadFileSystem(fsApt); } try { SerializationUtils.deserializeState(fs, path, state); return; } catch (IOException ioe) { if (ioe.getMessage().equals("Filesystem closed") && i < maxAttempts - 1) { continue; } else { throw ioe; } } } } public static StateStore<TaskState> openTaskStateStore(FileSystemJobStateful f) throws IOException { try (FileSystem fs = Help.loadFileSystem(f)) { return JobStateUtils.openTaskStateStore(Help.loadJobState(f, fs), fs); } } public static StateStore<TaskState> openTaskStateStore(FileSystemJobStateful js, FileSystem fs) throws IOException { return JobStateUtils.openTaskStateStoreUncached(loadJobState(js), fs); // public static StateStore<TaskState> openTaskStateStore(JobStateful js, FileSystem fs) throws IOException { // return JobStateUtils.openTaskStateStore(loadJobState(js, fs), fs); } private static void incrementJobStateAccess() { int numAccesses = jobStateAccessCount.getAndIncrement(); if (numAccesses % LOG_CACHE_STATS_EVERY_N_ACCESSES == 0) { log.info("JobState(numAccesses: {}) - {}", numAccesses, jobStateByPath.stats()); } } }
3,960
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work/styles/FileSystemApt.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.work.styles; import java.net.URI; import com.fasterxml.jackson.annotation.JsonIgnore; import org.apache.gobblin.configuration.State; /** Marks a type that can indicate a {@link org.apache.hadoop.fs.FileSystem} via its {@link URI} and configuration */ public interface FileSystemApt { URI getFileSystemUri(); @JsonIgnore // (because no-arg method resembles 'java bean property') State getFileSystemConfig(); }
3,961
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work/styles/FileSystemJobStateful.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.work.styles; /** Marks a type that can indicate both a {@link org.apache.hadoop.fs.FileSystem} and a {@link org.apache.gobblin.runtime.JobState} */ public interface FileSystemJobStateful extends JobStateful, FileSystemApt { }
3,962
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/work/styles/JobStateful.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.work.styles; import org.apache.hadoop.fs.Path; /** Marks a type that can indicate a {@link org.apache.gobblin.runtime.JobState} via its {@link Path} */ public interface JobStateful { Path getJobStatePath(); }
3,963
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/workflow/ProcessWorkUnitsWorkflow.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.workflow; import io.temporal.workflow.WorkflowInterface; import io.temporal.workflow.WorkflowMethod; import org.apache.gobblin.source.workunit.WorkUnit; import org.apache.gobblin.temporal.ddm.work.WUProcessingSpec; /** Temporal workflow for executing {@link WorkUnit}s to fulfill the work they specify. */ @WorkflowInterface public interface ProcessWorkUnitsWorkflow { /** @return the number of {@link WorkUnit}s cumulatively processed successfully */ @WorkflowMethod int process(WUProcessingSpec wuSpec); }
3,964
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/workflow
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/workflow/impl/NestingExecOfProcessWorkUnitWorkflowImpl.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.workflow.impl; import io.temporal.activity.ActivityOptions; import io.temporal.common.RetryOptions; import io.temporal.workflow.Async; import io.temporal.workflow.Promise; import io.temporal.workflow.Workflow; import java.time.Duration; import org.apache.gobblin.temporal.ddm.activity.ProcessWorkUnit; import org.apache.gobblin.temporal.ddm.work.WorkUnitClaimCheck; import org.apache.gobblin.temporal.util.nesting.workflow.AbstractNestingExecWorkflowImpl; /** {@link org.apache.gobblin.temporal.util.nesting.workflow.NestingExecWorkflow} for {@link ProcessWorkUnit} */ public class NestingExecOfProcessWorkUnitWorkflowImpl extends AbstractNestingExecWorkflowImpl<WorkUnitClaimCheck, Integer> { // RetryOptions specify how to automatically handle retries when Activities fail. private static final RetryOptions ACTIVITY_RETRY_OPTS = RetryOptions.newBuilder() .setInitialInterval(Duration.ofSeconds(3)) .setMaximumInterval(Duration.ofSeconds(100)) .setBackoffCoefficient(2) .setMaximumAttempts(4) .build(); private static final ActivityOptions ACTIVITY_OPTS = ActivityOptions.newBuilder() .setStartToCloseTimeout(Duration.ofSeconds(999)) .setRetryOptions(ACTIVITY_RETRY_OPTS) .build(); private final ProcessWorkUnit activityStub = Workflow.newActivityStub(ProcessWorkUnit.class, ACTIVITY_OPTS); @Override protected Promise<Integer> launchAsyncActivity(final WorkUnitClaimCheck wu) { return Async.function(activityStub::processWorkUnit, wu); } }
3,965
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/workflow
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/workflow/impl/ProcessWorkUnitsWorkflowImpl.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.workflow.impl; import java.util.Optional; import com.typesafe.config.ConfigFactory; import io.temporal.api.enums.v1.ParentClosePolicy; import io.temporal.workflow.ChildWorkflowOptions; import io.temporal.workflow.Workflow; import org.apache.gobblin.temporal.cluster.WorkerConfig; import org.apache.gobblin.temporal.ddm.work.WUProcessingSpec; import org.apache.gobblin.temporal.ddm.work.WorkUnitClaimCheck; import org.apache.gobblin.temporal.ddm.work.assistance.Help; import org.apache.gobblin.temporal.ddm.work.styles.FileSystemJobStateful; import org.apache.gobblin.temporal.ddm.workflow.ProcessWorkUnitsWorkflow; import org.apache.gobblin.temporal.ddm.work.EagerFsDirBackedWorkUnitClaimCheckWorkload; import org.apache.gobblin.temporal.util.nesting.work.WorkflowAddr; import org.apache.gobblin.temporal.util.nesting.work.Workload; import org.apache.gobblin.temporal.util.nesting.workflow.NestingExecWorkflow; public class ProcessWorkUnitsWorkflowImpl implements ProcessWorkUnitsWorkflow { public static final String CHILD_WORKFLOW_ID_BASE = "NestingExecWorkUnits"; @Override public int process(WUProcessingSpec workSpec) { Workload<WorkUnitClaimCheck> workload = createWorkload(workSpec); NestingExecWorkflow<WorkUnitClaimCheck> processingWorkflow = createProcessingWorkflow(workSpec); return processingWorkflow.performWorkload( WorkflowAddr.ROOT, workload, 0, workSpec.getTuning().getMaxBranchesPerTree(), workSpec.getTuning().getMaxSubTreesPerTree(), Optional.empty() ); } protected Workload<WorkUnitClaimCheck> createWorkload(WUProcessingSpec workSpec) { return new EagerFsDirBackedWorkUnitClaimCheckWorkload(workSpec.getFileSystemUri(), workSpec.getWorkUnitsDir()); } protected NestingExecWorkflow<WorkUnitClaimCheck> createProcessingWorkflow(FileSystemJobStateful f) { ChildWorkflowOptions childOpts = ChildWorkflowOptions.newBuilder() .setParentClosePolicy(ParentClosePolicy.PARENT_CLOSE_POLICY_ABANDON) .setWorkflowId(Help.qualifyNamePerExec(CHILD_WORKFLOW_ID_BASE, f, WorkerConfig.of(this).orElse(ConfigFactory.empty()))) .build(); // TODO: to incorporate multiple different concrete `NestingExecWorkflow` sub-workflows in the same super-workflow... shall we use queues?!?!? return Workflow.newChildWorkflowStub(NestingExecWorkflow.class, childOpts); } }
3,966
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/ddm/worker/WorkFulfillmentWorker.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.ddm.worker; import java.util.concurrent.TimeUnit; import com.typesafe.config.Config; import io.temporal.client.WorkflowClient; import io.temporal.worker.WorkerOptions; import org.apache.gobblin.temporal.cluster.AbstractTemporalWorker; import org.apache.gobblin.temporal.ddm.activity.impl.ProcessWorkUnitImpl; import org.apache.gobblin.temporal.ddm.workflow.impl.NestingExecOfProcessWorkUnitWorkflowImpl; import org.apache.gobblin.temporal.ddm.workflow.impl.ProcessWorkUnitsWorkflowImpl; /** Worker for the {@link ProcessWorkUnitsWorkflowImpl} super-workflow */ public class WorkFulfillmentWorker extends AbstractTemporalWorker { public static final long DEADLOCK_DETECTION_TIMEOUT_SECONDS = 120; public static final int MAX_EXECUTION_CONCURRENCY = 3; public WorkFulfillmentWorker(Config config, WorkflowClient workflowClient) { super(config, workflowClient); } @Override protected Class<?>[] getWorkflowImplClasses() { return new Class[] { ProcessWorkUnitsWorkflowImpl.class, NestingExecOfProcessWorkUnitWorkflowImpl.class }; } @Override protected Object[] getActivityImplInstances() { return new Object[] { new ProcessWorkUnitImpl() }; } @Override protected WorkerOptions createWorkerOptions() { return WorkerOptions.newBuilder() // default is only 1s - WAY TOO SHORT for `o.a.hadoop.fs.FileSystem#listStatus`! .setDefaultDeadlockDetectionTimeout(TimeUnit.SECONDS.toMillis(DEADLOCK_DETECTION_TIMEOUT_SECONDS)) .setMaxConcurrentActivityExecutionSize(MAX_EXECUTION_CONCURRENCY) .setMaxConcurrentLocalActivityExecutionSize(MAX_EXECUTION_CONCURRENCY) .setMaxConcurrentWorkflowTaskExecutionSize(MAX_EXECUTION_CONCURRENCY) .build(); } }
3,967
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows/helloworld/FormatActivityImpl.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.workflows.helloworld; public class FormatActivityImpl implements FormatActivity { @Override public String composeGreeting(String name) { return "Hello " + name + "!"; } }
3,968
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows/helloworld/GreetingWorkflowImpl.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.workflows.helloworld; import java.time.Duration; import io.temporal.activity.ActivityOptions; import io.temporal.workflow.Workflow; public class GreetingWorkflowImpl implements GreetingWorkflow { /* * At least one of the following options needs to be defined: * - setStartToCloseTimeout * - setScheduleToCloseTimeout */ ActivityOptions options = ActivityOptions.newBuilder() .setStartToCloseTimeout(Duration.ofSeconds(60)) .build(); /* * Define the HelloWorldActivity stub. Activity stubs are proxies for activity invocations that * are executed outside of the workflow thread on the activity worker, that can be on a * different host. Temporal is going to dispatch the activity results back to the workflow and * unblock the stub as soon as activity is completed on the activity worker. * * The activity options that were defined above are passed in as a parameter. */ private final FormatActivity activity = Workflow.newActivityStub(FormatActivity.class, options); // This is the entry point to the Workflow. @Override public String getGreeting(String name) { /** * If there were other Activity methods they would be orchestrated here or from within other Activities. * This is a blocking call that returns only after the activity has completed. */ return activity.composeGreeting(name); } }
3,969
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows/helloworld/HelloWorldWorker.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.workflows.helloworld; import com.typesafe.config.Config; import io.temporal.client.WorkflowClient; import org.apache.gobblin.temporal.cluster.AbstractTemporalWorker; public class HelloWorldWorker extends AbstractTemporalWorker { public HelloWorldWorker(Config config, WorkflowClient workflowClient) { super(config, workflowClient); } @Override protected Class<?>[] getWorkflowImplClasses() { return new Class[] { GreetingWorkflowImpl.class }; } @Override protected Object[] getActivityImplInstances() { return new Object[] { new FormatActivityImpl() }; } }
3,970
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows/helloworld/HelloWorldJobLauncher.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.workflows.helloworld; import java.util.List; import java.util.Properties; import java.util.concurrent.ConcurrentHashMap; import org.apache.hadoop.fs.Path; import io.temporal.client.WorkflowOptions; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.metrics.Tag; import org.apache.gobblin.runtime.JobLauncher; import org.apache.gobblin.source.workunit.WorkUnit; import org.apache.gobblin.temporal.cluster.GobblinTemporalTaskRunner; import org.apache.gobblin.temporal.joblauncher.GobblinTemporalJobLauncher; import org.apache.gobblin.temporal.joblauncher.GobblinTemporalJobScheduler; /** * An implementation of {@link JobLauncher} that launches a Gobblin job using the Temporal task framework. * * <p> * This class is instantiated by the {@link GobblinTemporalJobScheduler#buildJobLauncher(Properties)} on every job submission to launch the Gobblin job. * The actual task execution happens in the {@link GobblinTemporalTaskRunner}, usually in a different process. * </p> */ @Alpha @Slf4j public class HelloWorldJobLauncher extends GobblinTemporalJobLauncher { public HelloWorldJobLauncher(Properties jobProps, Path appWorkDir, List<? extends Tag<?>> metadataTags, ConcurrentHashMap<String, Boolean> runningMap) throws Exception { super(jobProps, appWorkDir, metadataTags, runningMap); } @Override public void submitJob(List<WorkUnit> workunits) { WorkflowOptions options = WorkflowOptions.newBuilder().setTaskQueue(queueName).build(); GreetingWorkflow greetingWorkflow = this.client.newWorkflowStub(GreetingWorkflow.class, options); greetingWorkflow.getGreeting("Gobblin"); } }
3,971
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows/helloworld/FormatActivity.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.workflows.helloworld; import io.temporal.activity.ActivityInterface; @ActivityInterface public interface FormatActivity { // Define your activity methods which can be called during workflow execution String composeGreeting(String name); }
3,972
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows/helloworld/GreetingWorkflow.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.workflows.helloworld; import io.temporal.workflow.WorkflowInterface; import io.temporal.workflow.WorkflowMethod; @WorkflowInterface public interface GreetingWorkflow { /** * This is the method that is executed when the Workflow Execution is started. The Workflow * Execution completes when this method finishes execution. */ @WorkflowMethod String getGreeting(String name); }
3,973
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/workflows/client/TemporalWorkflowClientFactory.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.workflows.client; import java.io.ByteArrayInputStream; import java.io.File; import java.io.IOException; import java.io.InputStream; import java.security.KeyStore; import java.util.Arrays; import java.util.Collections; import java.util.List; import org.apache.commons.io.FileUtils; import com.typesafe.config.Config; import com.typesafe.config.ConfigFactory; import io.grpc.netty.shaded.io.grpc.netty.GrpcSslContexts; import io.grpc.netty.shaded.io.netty.handler.ssl.SslContext; import io.temporal.client.WorkflowClient; import io.temporal.client.WorkflowClientOptions; import io.temporal.serviceclient.WorkflowServiceStubs; import io.temporal.serviceclient.WorkflowServiceStubsOptions; import javax.net.ssl.KeyManagerFactory; import javax.net.ssl.TrustManagerFactory; import org.apache.gobblin.cluster.GobblinClusterUtils; public class TemporalWorkflowClientFactory { public static WorkflowServiceStubs createServiceInstance(String connectionUri) throws Exception { GobblinClusterUtils.setSystemProperties(ConfigFactory.load()); Config config = GobblinClusterUtils.addDynamicConfig(ConfigFactory.load()); String SHARED_KAFKA_CONFIG_PREFIX_WITH_DOT = "gobblin.kafka.sharedConfig."; String SSL_KEYMANAGER_ALGORITHM = SHARED_KAFKA_CONFIG_PREFIX_WITH_DOT + "ssl.keymanager.algorithm"; String SSL_KEYSTORE_TYPE = SHARED_KAFKA_CONFIG_PREFIX_WITH_DOT + "ssl.keystore.type"; String SSL_KEYSTORE_LOCATION = SHARED_KAFKA_CONFIG_PREFIX_WITH_DOT + "ssl.keystore.location"; String SSL_KEY_PASSWORD = SHARED_KAFKA_CONFIG_PREFIX_WITH_DOT + "ssl.key.password"; String SSL_TRUSTSTORE_LOCATION = SHARED_KAFKA_CONFIG_PREFIX_WITH_DOT + "ssl.truststore.location"; String SSL_TRUSTSTORE_PASSWORD = SHARED_KAFKA_CONFIG_PREFIX_WITH_DOT + "ssl.truststore.password"; List<String> SSL_CONFIG_DEFAULT_SSL_PROTOCOLS = Collections.unmodifiableList( Arrays.asList("TLSv1.2")); List<String> SSL_CONFIG_DEFAULT_CIPHER_SUITES = Collections.unmodifiableList(Arrays.asList( // The following list is from https://github.com/netty/netty/blob/4.1/codec-http2/src/main/java/io/netty/handler/codec/http2/Http2SecurityUtil.java#L50 "TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256", /* REQUIRED BY HTTP/2 SPEC */ "TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256", /* REQUIRED BY HTTP/2 SPEC */ "TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384", "TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384", "TLS_ECDHE_ECDSA_WITH_CHACHA20_POLY1305_SHA256", "TLS_ECDHE_RSA_WITH_CHACHA20_POLY1305_SHA256" )); String keyStoreType = config.getString(SSL_KEYSTORE_TYPE); File keyStoreFile = new File(config.getString(SSL_KEYSTORE_LOCATION)); String keyStorePassword = config.getString(SSL_KEY_PASSWORD); KeyStore keyStore = KeyStore.getInstance(keyStoreType); keyStore.load(toInputStream(keyStoreFile), keyStorePassword.toCharArray()); // Set key manager from key store String sslKeyManagerAlgorithm = config.getString(SSL_KEYMANAGER_ALGORITHM); KeyManagerFactory keyManagerFactory = KeyManagerFactory.getInstance(sslKeyManagerAlgorithm); keyManagerFactory.init(keyStore, keyStorePassword.toCharArray()); // Set trust manager from trust store KeyStore trustStore = KeyStore.getInstance("JKS"); File trustStoreFile = new File(config.getString(SSL_TRUSTSTORE_LOCATION)); String trustStorePassword = config.getString(SSL_TRUSTSTORE_PASSWORD); trustStore.load(toInputStream(trustStoreFile), trustStorePassword.toCharArray()); TrustManagerFactory trustManagerFactory = TrustManagerFactory.getInstance("SunX509"); trustManagerFactory.init(trustStore); SslContext sslContext = GrpcSslContexts.forClient() .keyManager(keyManagerFactory) .trustManager(trustManagerFactory) .protocols(SSL_CONFIG_DEFAULT_SSL_PROTOCOLS) .ciphers(SSL_CONFIG_DEFAULT_CIPHER_SUITES) .build(); WorkflowServiceStubsOptions options = WorkflowServiceStubsOptions.newBuilder() .setTarget(connectionUri) .setEnableHttps(true) .setSslContext(sslContext) .build(); return WorkflowServiceStubs.newServiceStubs(options); } public static WorkflowClient createClientInstance(WorkflowServiceStubs service, String namespace) { WorkflowClientOptions options = WorkflowClientOptions.newBuilder().setNamespace(namespace).build(); return WorkflowClient.newInstance(service, options); } private static InputStream toInputStream(File storeFile) throws IOException { byte[] data = FileUtils.readFileToByteArray(storeFile); return new ByteArrayInputStream(data); } }
3,974
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/yarn/YarnTemporalAppMasterSecurityManager.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.yarn; import com.google.common.base.Throwables; import com.google.common.eventbus.EventBus; import com.typesafe.config.Config; import java.io.IOException; import org.apache.gobblin.util.logs.LogCopier; import org.apache.gobblin.yarn.YarnContainerSecurityManager; import org.apache.gobblin.yarn.event.DelegationTokenUpdatedEvent; import org.apache.hadoop.fs.FileSystem; /** * Copied from {@link org.apache.gobblin.yarn.YarnAppMasterSecurityManager} that uses the {@link YarnService} * * This class was created for a fast way to start building out Gobblin on temporal without affecting mainline Yarn/Helix code */ public class YarnTemporalAppMasterSecurityManager extends YarnContainerSecurityManager { private YarnService _yarnService; public YarnTemporalAppMasterSecurityManager(Config config, FileSystem fs, EventBus eventBus, LogCopier logCopier, YarnService yarnService) { super(config, fs, eventBus, logCopier); this._yarnService = yarnService; } @Override public void handleTokenFileUpdatedEvent(DelegationTokenUpdatedEvent delegationTokenUpdatedEvent) { super.handleTokenFileUpdatedEvent(delegationTokenUpdatedEvent); try { _yarnService.updateToken(); } catch (IOException ioe) { throw Throwables.propagate(ioe); } } }
3,975
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/yarn/YarnService.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.yarn; import com.google.common.annotations.VisibleForTesting; import com.google.common.base.Optional; import com.google.common.base.Preconditions; import com.google.common.base.Strings; import com.google.common.base.Throwables; import com.google.common.cache.Cache; import com.google.common.cache.CacheBuilder; import com.google.common.collect.ImmutableList; import com.google.common.collect.ImmutableMap; import com.google.common.collect.Maps; import com.google.common.eventbus.EventBus; import com.google.common.eventbus.Subscribe; import com.google.common.io.Closer; import com.google.common.util.concurrent.AbstractIdleService; import com.typesafe.config.Config; import java.io.File; import java.io.IOException; import java.nio.ByteBuffer; import java.time.Duration; import java.util.Arrays; import java.util.Collection; import java.util.Collections; import java.util.HashMap; import java.util.Iterator; import java.util.List; import java.util.Map; import java.util.Set; import java.util.concurrent.ConcurrentHashMap; import java.util.concurrent.ConcurrentMap; import java.util.concurrent.ExecutorService; import java.util.concurrent.TimeUnit; import java.util.concurrent.atomic.AtomicInteger; import java.util.stream.IntStream; import lombok.AccessLevel; import lombok.Getter; import org.slf4j.Logger; import org.slf4j.LoggerFactory; import org.apache.gobblin.cluster.GobblinClusterConfigurationKeys; import org.apache.gobblin.cluster.GobblinClusterMetricTagNames; import org.apache.gobblin.cluster.GobblinClusterUtils; import org.apache.gobblin.cluster.event.ClusterManagerShutdownRequest; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.metrics.GobblinMetrics; import org.apache.gobblin.metrics.MetricReporterException; import org.apache.gobblin.metrics.MultiReporterException; import org.apache.gobblin.metrics.Tag; import org.apache.gobblin.metrics.event.EventSubmitter; import org.apache.gobblin.util.ConfigUtils; import org.apache.gobblin.util.ExecutorsUtils; import org.apache.gobblin.util.JvmUtils; import org.apache.gobblin.util.executors.ScalingThreadPoolExecutor; import org.apache.gobblin.yarn.GobblinYarnConfigurationKeys; import org.apache.gobblin.yarn.GobblinYarnEventConstants; import org.apache.gobblin.yarn.GobblinYarnMetricTagNames; import org.apache.gobblin.yarn.YarnHelixUtils; import org.apache.gobblin.yarn.event.ContainerReleaseRequest; import org.apache.gobblin.yarn.event.ContainerShutdownRequest; import org.apache.gobblin.yarn.event.NewContainerRequest; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.FileStatus; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.apache.hadoop.io.DataOutputBuffer; import org.apache.hadoop.security.Credentials; import org.apache.hadoop.security.UserGroupInformation; import org.apache.hadoop.security.token.Token; import org.apache.hadoop.yarn.api.ApplicationConstants; import org.apache.hadoop.yarn.api.protocolrecords.RegisterApplicationMasterResponse; import org.apache.hadoop.yarn.api.records.ApplicationAccessType; import org.apache.hadoop.yarn.api.records.Container; import org.apache.hadoop.yarn.api.records.ContainerExitStatus; import org.apache.hadoop.yarn.api.records.ContainerId; import org.apache.hadoop.yarn.api.records.ContainerLaunchContext; import org.apache.hadoop.yarn.api.records.ContainerStatus; import org.apache.hadoop.yarn.api.records.FinalApplicationStatus; import org.apache.hadoop.yarn.api.records.LocalResource; import org.apache.hadoop.yarn.api.records.LocalResourceType; import org.apache.hadoop.yarn.api.records.NodeReport; import org.apache.hadoop.yarn.api.records.Priority; import org.apache.hadoop.yarn.api.records.Resource; import org.apache.hadoop.yarn.api.records.ResourceRequest; import org.apache.hadoop.yarn.client.api.AMRMClient; import org.apache.hadoop.yarn.client.api.async.AMRMClientAsync; import org.apache.hadoop.yarn.client.api.async.NMClientAsync; import org.apache.hadoop.yarn.conf.YarnConfiguration; import org.apache.hadoop.yarn.exceptions.YarnException; import org.apache.hadoop.yarn.security.AMRMTokenIdentifier; import org.apache.hadoop.yarn.util.Records; /** * This class is responsible for all Yarn-related stuffs including ApplicationMaster registration, * ApplicationMaster un-registration, Yarn container management, etc. * * NOTE: This is a stripped down version of {@link org.apache.gobblin.yarn.YarnService} that is used for temporal testing * without any dependency on Helix. There are some references to helix concepts, but they are left in for the sake of * keeping some features in-tact. They don't have an actual dependency on helix anymore. * */ class YarnService extends AbstractIdleService { private static final Logger LOGGER = LoggerFactory.getLogger(YarnService.class); private static final String UNKNOWN_HELIX_INSTANCE = "UNKNOWN"; private final String applicationName; private final String applicationId; private final String appViewAcl; //Default helix instance tag derived from cluster level config private final String helixInstanceTags; private final Config config; private final EventBus eventBus; private final Configuration yarnConfiguration; private final FileSystem fs; private final Optional<GobblinMetrics> gobblinMetrics; private final Optional<EventSubmitter> eventSubmitter; @VisibleForTesting @Getter(AccessLevel.PROTECTED) private final AMRMClientAsync<AMRMClient.ContainerRequest> amrmClientAsync; private final NMClientAsync nmClientAsync; private final ExecutorService containerLaunchExecutor; private final int initialContainers; private final int requestedContainerMemoryMbs; private final int requestedContainerCores; private final int jvmMemoryOverheadMbs; private final double jvmMemoryXmxRatio; private final boolean containerHostAffinityEnabled; private final int helixInstanceMaxRetries; private final Optional<String> containerJvmArgs; private final String containerTimezone; @Getter(AccessLevel.PROTECTED) private volatile Optional<Resource> maxResourceCapacity = Optional.absent(); // Security tokens for accessing HDFS private ByteBuffer tokens; private final Closer closer = Closer.create(); private final Object allContainersStopped = new Object(); // A map from container IDs to Container instances, Helix participant IDs of the containers and Helix Tag @VisibleForTesting @Getter(AccessLevel.PROTECTED) private final ConcurrentMap<ContainerId, ContainerInfo> containerMap = Maps.newConcurrentMap(); // A cache of the containers with an outstanding container release request. // This is a cache instead of a set to get the automatic cleanup in case a container completes before the requested // release. @VisibleForTesting @Getter(AccessLevel.PROTECTED) private final Cache<ContainerId, String> releasedContainerCache; // A map from Helix instance names to the number times the instances are retried to be started private final ConcurrentMap<String, AtomicInteger> helixInstanceRetryCount = Maps.newConcurrentMap(); // A concurrent HashSet of unused Helix instance names. An unused Helix instance name gets put // into the set if the container running the instance completes. Unused Helix // instance names get picked up when replacement containers get allocated. private final Set<String> unusedHelixInstanceNames = ConcurrentHashMap.newKeySet(); // The map from helix tag to allocated container count private final ConcurrentMap<String, AtomicInteger> allocatedContainerCountMap = Maps.newConcurrentMap(); private final ConcurrentMap<ContainerId, String> removedContainerID = Maps.newConcurrentMap(); private final AtomicInteger priorityNumGenerator = new AtomicInteger(0); private final Map<String, Integer> resourcePriorityMap = new HashMap<>(); private volatile boolean shutdownInProgress = false; public YarnService(Config config, String applicationName, String applicationId, YarnConfiguration yarnConfiguration, FileSystem fs, EventBus eventBus) throws Exception { this.applicationName = applicationName; this.applicationId = applicationId; this.config = config; this.eventBus = eventBus; this.gobblinMetrics = config.getBoolean(ConfigurationKeys.METRICS_ENABLED_KEY) ? Optional.of(buildGobblinMetrics()) : Optional.<GobblinMetrics>absent(); this.eventSubmitter = config.getBoolean(ConfigurationKeys.METRICS_ENABLED_KEY) ? Optional.of(buildEventSubmitter()) : Optional.<EventSubmitter>absent(); this.yarnConfiguration = yarnConfiguration; this.fs = fs; int amRmHeartbeatIntervalMillis = Long.valueOf(TimeUnit.SECONDS.toMillis( ConfigUtils.getInt(config, GobblinYarnConfigurationKeys.AMRM_HEARTBEAT_INTERVAL_SECS, GobblinYarnConfigurationKeys.DEFAULT_AMRM_HEARTBEAT_INTERVAL_SECS))).intValue(); this.amrmClientAsync = closer.register( AMRMClientAsync.createAMRMClientAsync(amRmHeartbeatIntervalMillis, new AMRMClientCallbackHandler())); this.amrmClientAsync.init(this.yarnConfiguration); this.nmClientAsync = closer.register(NMClientAsync.createNMClientAsync(getNMClientCallbackHandler())); this.nmClientAsync.init(this.yarnConfiguration); this.initialContainers = config.getInt(GobblinYarnConfigurationKeys.INITIAL_CONTAINERS_KEY); this.requestedContainerMemoryMbs = config.getInt(GobblinYarnConfigurationKeys.CONTAINER_MEMORY_MBS_KEY); this.requestedContainerCores = config.getInt(GobblinYarnConfigurationKeys.CONTAINER_CORES_KEY); this.containerHostAffinityEnabled = config.getBoolean(GobblinYarnConfigurationKeys.CONTAINER_HOST_AFFINITY_ENABLED); this.helixInstanceMaxRetries = config.getInt(GobblinYarnConfigurationKeys.HELIX_INSTANCE_MAX_RETRIES); this.helixInstanceTags = ConfigUtils.getString(config, GobblinClusterConfigurationKeys.HELIX_INSTANCE_TAGS_KEY, GobblinClusterConfigurationKeys.HELIX_DEFAULT_TAG); this.containerJvmArgs = config.hasPath(GobblinYarnConfigurationKeys.CONTAINER_JVM_ARGS_KEY) ? Optional.of(config.getString(GobblinYarnConfigurationKeys.CONTAINER_JVM_ARGS_KEY)) : Optional.<String>absent(); int numContainerLaunchThreads = ConfigUtils.getInt(config, GobblinYarnConfigurationKeys.MAX_CONTAINER_LAUNCH_THREADS_KEY, GobblinYarnConfigurationKeys.DEFAULT_MAX_CONTAINER_LAUNCH_THREADS); this.containerLaunchExecutor = ScalingThreadPoolExecutor.newScalingThreadPool(5, numContainerLaunchThreads, 0L, ExecutorsUtils.newThreadFactory(Optional.of(LOGGER), Optional.of("ContainerLaunchExecutor"))); this.tokens = getSecurityTokens(); this.releasedContainerCache = CacheBuilder.newBuilder().expireAfterAccess(ConfigUtils.getInt(config, GobblinYarnConfigurationKeys.RELEASED_CONTAINERS_CACHE_EXPIRY_SECS, GobblinYarnConfigurationKeys.DEFAULT_RELEASED_CONTAINERS_CACHE_EXPIRY_SECS), TimeUnit.SECONDS).build(); this.jvmMemoryXmxRatio = ConfigUtils.getDouble(this.config, GobblinYarnConfigurationKeys.CONTAINER_JVM_MEMORY_XMX_RATIO_KEY, GobblinYarnConfigurationKeys.DEFAULT_CONTAINER_JVM_MEMORY_XMX_RATIO); Preconditions.checkArgument(this.jvmMemoryXmxRatio >= 0 && this.jvmMemoryXmxRatio <= 1, GobblinYarnConfigurationKeys.CONTAINER_JVM_MEMORY_XMX_RATIO_KEY + " must be between 0 and 1 inclusive"); this.jvmMemoryOverheadMbs = ConfigUtils.getInt(this.config, GobblinYarnConfigurationKeys.CONTAINER_JVM_MEMORY_OVERHEAD_MBS_KEY, GobblinYarnConfigurationKeys.DEFAULT_CONTAINER_JVM_MEMORY_OVERHEAD_MBS); Preconditions.checkArgument(this.jvmMemoryOverheadMbs < this.requestedContainerMemoryMbs * this.jvmMemoryXmxRatio, GobblinYarnConfigurationKeys.CONTAINER_JVM_MEMORY_OVERHEAD_MBS_KEY + " cannot be more than " + GobblinYarnConfigurationKeys.CONTAINER_MEMORY_MBS_KEY + " * " + GobblinYarnConfigurationKeys.CONTAINER_JVM_MEMORY_XMX_RATIO_KEY); this.appViewAcl = ConfigUtils.getString(this.config, GobblinYarnConfigurationKeys.APP_VIEW_ACL, GobblinYarnConfigurationKeys.DEFAULT_APP_VIEW_ACL); this.containerTimezone = ConfigUtils.getString(this.config, GobblinYarnConfigurationKeys.GOBBLIN_YARN_CONTAINER_TIMEZONE, GobblinYarnConfigurationKeys.DEFAULT_GOBBLIN_YARN_CONTAINER_TIMEZONE); } @SuppressWarnings("unused") @Subscribe public void handleNewContainerRequest(NewContainerRequest newContainerRequest) { if (!this.maxResourceCapacity.isPresent()) { LOGGER.error(String.format( "Unable to handle new container request as maximum resource capacity is not available: " + "[memory (MBs) requested = %d, vcores requested = %d]", this.requestedContainerMemoryMbs, this.requestedContainerCores)); return; } requestContainer(newContainerRequest.getReplacedContainer().transform(container -> container.getNodeId().getHost()), newContainerRequest.getResource()); } protected NMClientCallbackHandler getNMClientCallbackHandler() { return new NMClientCallbackHandler(); } @SuppressWarnings("unused") @Subscribe public void handleContainerShutdownRequest(ContainerShutdownRequest containerShutdownRequest) { for (Container container : containerShutdownRequest.getContainers()) { LOGGER.info(String.format("Stopping container %s running on %s", container.getId(), container.getNodeId())); this.nmClientAsync.stopContainerAsync(container.getId(), container.getNodeId()); } } /** * Request the Resource Manager to release the container * @param containerReleaseRequest containers to release */ @Subscribe public void handleContainerReleaseRequest(ContainerReleaseRequest containerReleaseRequest) { for (Container container : containerReleaseRequest.getContainers()) { LOGGER.info(String.format("Releasing container %s running on %s", container.getId(), container.getNodeId())); // Record that this container was explicitly released so that a new one is not spawned to replace it // Put the container id in the releasedContainerCache before releasing it so that handleContainerCompletion() // can check for the container id and skip spawning a replacement container. // Note that this is the best effort since these are asynchronous operations and a container may abort concurrently // with the release call. So in some cases a replacement container may have already been spawned before // the container is put into the black list. this.releasedContainerCache.put(container.getId(), ""); this.amrmClientAsync.releaseAssignedContainer(container.getId()); } } @Override protected synchronized void startUp() throws Exception { LOGGER.info("Starting the TemporalYarnService"); // Register itself with the EventBus for container-related requests this.eventBus.register(this); this.amrmClientAsync.start(); this.nmClientAsync.start(); // The ApplicationMaster registration response is used to determine the maximum resource capacity of the cluster RegisterApplicationMasterResponse response = this.amrmClientAsync.registerApplicationMaster( GobblinClusterUtils.getHostname(), -1, ""); LOGGER.info("ApplicationMaster registration response: " + response); this.maxResourceCapacity = Optional.of(response.getMaximumResourceCapability()); LOGGER.info("Requesting initial containers"); requestInitialContainers(this.initialContainers); } @Override protected void shutDown() throws IOException { LOGGER.info("Stopping the TemporalYarnService"); this.shutdownInProgress = true; try { ExecutorsUtils.shutdownExecutorService(this.containerLaunchExecutor, Optional.of(LOGGER)); // Stop the running containers for (ContainerInfo containerInfo : this.containerMap.values()) { LOGGER.info("Stopping container {} running participant {}", containerInfo.getContainer().getId(), containerInfo.getHelixParticipantId()); this.nmClientAsync.stopContainerAsync(containerInfo.getContainer().getId(), containerInfo.getContainer().getNodeId()); } if (!this.containerMap.isEmpty()) { synchronized (this.allContainersStopped) { try { // Wait 5 minutes for the containers to stop Duration waitTimeout = Duration.ofMinutes(5); this.allContainersStopped.wait(waitTimeout.toMillis()); LOGGER.info("All of the containers have been stopped"); } catch (InterruptedException ie) { Thread.currentThread().interrupt(); } } } this.amrmClientAsync.unregisterApplicationMaster(FinalApplicationStatus.SUCCEEDED, null, null); } catch (IOException | YarnException e) { LOGGER.error("Failed to unregister the ApplicationMaster", e); } finally { try { this.closer.close(); } finally { if (this.gobblinMetrics.isPresent()) { this.gobblinMetrics.get().stopMetricsReporting(); } } } } public void updateToken() throws IOException{ this.tokens = getSecurityTokens(); } private GobblinMetrics buildGobblinMetrics() { // Create tags list ImmutableList.Builder<Tag<?>> tags = new ImmutableList.Builder<>(); tags.add(new Tag<>(GobblinClusterMetricTagNames.APPLICATION_ID, this.applicationId)); tags.add(new Tag<>(GobblinClusterMetricTagNames.APPLICATION_NAME, this.applicationName)); // Intialize Gobblin metrics and start reporters GobblinMetrics gobblinMetrics = GobblinMetrics.get(this.applicationId, null, tags.build()); try { gobblinMetrics.startMetricReporting(ConfigUtils.configToProperties(config)); } catch (MultiReporterException ex) { for (MetricReporterException e: ex.getExceptions()) { LOGGER.error("Failed to start {} {} reporter.", e.getSinkType().name(), e.getReporterType().name(), e); } } return gobblinMetrics; } private EventSubmitter buildEventSubmitter() { return new EventSubmitter.Builder(this.gobblinMetrics.get().getMetricContext(), GobblinYarnEventConstants.EVENT_NAMESPACE) .build(); } /** * Request an allocation of containers. If numTargetContainers is larger than the max of current and expected number * of containers then additional containers are requested. * <p> * If numTargetContainers is less than the current number of allocated containers then release free containers. * Shrinking is relative to the number of currently allocated containers since it takes time for containers * to be allocated and assigned work and we want to avoid releasing a container prematurely before it is assigned * work. This means that a container may not be released even though numTargetContainers is less than the requested * number of containers. The intended usage is for the caller of this method to make periodic calls to attempt to * adjust the cluster towards the desired number of containers. * * @param inUseInstances a set of in use instances * @return whether successfully requested the target number of containers */ public synchronized boolean requestTargetNumberOfContainers(int numContainers, Set<String> inUseInstances) { int defaultContainerMemoryMbs = config.getInt(GobblinYarnConfigurationKeys.CONTAINER_MEMORY_MBS_KEY); int defaultContainerCores = config.getInt(GobblinYarnConfigurationKeys. CONTAINER_CORES_KEY); LOGGER.info("Trying to set numTargetContainers={}, in-use helix instances count is {}, container map size is {}", numContainers, inUseInstances.size(), this.containerMap.size()); requestContainers(numContainers, Resource.newInstance(defaultContainerMemoryMbs, defaultContainerCores)); LOGGER.info("Current tag-container desired count:{}, tag-container allocated: {}", numContainers, this.allocatedContainerCountMap); return true; } // Request initial containers with default resource and helix tag private void requestInitialContainers(int containersRequested) { requestTargetNumberOfContainers(containersRequested, Collections.EMPTY_SET); } private void requestContainer(Optional<String> preferredNode, Optional<Resource> resourceOptional) { Resource desiredResource = resourceOptional.or(Resource.newInstance( this.requestedContainerMemoryMbs, this.requestedContainerCores)); requestContainer(preferredNode, desiredResource); } /** * Request {@param numContainers} from yarn with the specified resource. Resources will be allocated without a preferred * node * @param numContainers * @param resource */ private void requestContainers(int numContainers, Resource resource) { LOGGER.info("Requesting {} containers with resource={}", numContainers, resource); IntStream.range(0, numContainers) .forEach(i -> requestContainer(Optional.absent(), resource)); } // Request containers with specific resource requirement private void requestContainer(Optional<String> preferredNode, Resource resource) { // Fail if Yarn cannot meet container resource requirements Preconditions.checkArgument(resource.getMemory() <= this.maxResourceCapacity.get().getMemory() && resource.getVirtualCores() <= this.maxResourceCapacity.get().getVirtualCores(), "Resource requirement must less than the max resource capacity. Requested resource" + resource.toString() + " exceed the max resource limit " + this.maxResourceCapacity.get().toString()); // Due to YARN-314, different resource capacity needs different priority, otherwise Yarn will not allocate container Priority priority = Records.newRecord(Priority.class); if(!resourcePriorityMap.containsKey(resource.toString())) { resourcePriorityMap.put(resource.toString(), priorityNumGenerator.getAndIncrement()); } int priorityNum = resourcePriorityMap.get(resource.toString()); priority.setPriority(priorityNum); String[] preferredNodes = preferredNode.isPresent() ? new String[] {preferredNode.get()} : null; this.amrmClientAsync.addContainerRequest( new AMRMClient.ContainerRequest(resource, preferredNodes, null, priority)); } protected ContainerLaunchContext newContainerLaunchContext(ContainerInfo containerInfo) throws IOException { Path appWorkDir = GobblinClusterUtils.getAppWorkDirPathFromConfig(this.config, this.fs, this.applicationName, this.applicationId); Path containerWorkDir = new Path(appWorkDir, GobblinYarnConfigurationKeys.CONTAINER_WORK_DIR_NAME); Map<String, LocalResource> resourceMap = Maps.newHashMap(); addContainerLocalResources(new Path(appWorkDir, GobblinYarnConfigurationKeys.LIB_JARS_DIR_NAME), resourceMap); addContainerLocalResources(new Path(containerWorkDir, GobblinYarnConfigurationKeys.APP_JARS_DIR_NAME), resourceMap); addContainerLocalResources( new Path(containerWorkDir, GobblinYarnConfigurationKeys.APP_FILES_DIR_NAME), resourceMap); if (this.config.hasPath(GobblinYarnConfigurationKeys.CONTAINER_FILES_REMOTE_KEY)) { YarnHelixUtils.addRemoteFilesToLocalResources(this.config.getString(GobblinYarnConfigurationKeys.CONTAINER_FILES_REMOTE_KEY), resourceMap, yarnConfiguration); } if (this.config.hasPath(GobblinYarnConfigurationKeys.CONTAINER_ZIPS_REMOTE_KEY)) { YarnHelixUtils.addRemoteZipsToLocalResources(this.config.getString(GobblinYarnConfigurationKeys.CONTAINER_ZIPS_REMOTE_KEY), resourceMap, yarnConfiguration); } ContainerLaunchContext containerLaunchContext = Records.newRecord(ContainerLaunchContext.class); containerLaunchContext.setLocalResources(resourceMap); containerLaunchContext.setEnvironment(YarnHelixUtils.getEnvironmentVariables(this.yarnConfiguration)); containerLaunchContext.setCommands(Arrays.asList(containerInfo.getStartupCommand())); Map<ApplicationAccessType, String> acls = new HashMap<>(1); acls.put(ApplicationAccessType.VIEW_APP, this.appViewAcl); containerLaunchContext.setApplicationACLs(acls); if (UserGroupInformation.isSecurityEnabled()) { containerLaunchContext.setTokens(this.tokens.duplicate()); } return containerLaunchContext; } private void addContainerLocalResources(Path destDir, Map<String, LocalResource> resourceMap) throws IOException { if (!this.fs.exists(destDir)) { LOGGER.warn(String.format("Path %s does not exist so no container LocalResource to add", destDir)); return; } FileStatus[] statuses = this.fs.listStatus(destDir); if (statuses != null) { for (FileStatus status : statuses) { YarnHelixUtils.addFileAsLocalResource(this.fs, status.getPath(), LocalResourceType.FILE, resourceMap); } } } protected ByteBuffer getSecurityTokens() throws IOException { Credentials credentials = UserGroupInformation.getCurrentUser().getCredentials(); Closer closer = Closer.create(); try { DataOutputBuffer dataOutputBuffer = closer.register(new DataOutputBuffer()); credentials.writeTokenStorageToStream(dataOutputBuffer); // Remove the AM->RM token so that containers cannot access it Iterator<Token<?>> tokenIterator = credentials.getAllTokens().iterator(); while (tokenIterator.hasNext()) { Token<?> token = tokenIterator.next(); if (token.getKind().equals(AMRMTokenIdentifier.KIND_NAME)) { tokenIterator.remove(); } } return ByteBuffer.wrap(dataOutputBuffer.getData(), 0, dataOutputBuffer.getLength()); } catch (Throwable t) { throw closer.rethrow(t); } finally { closer.close(); } } @VisibleForTesting protected String buildContainerCommand(Container container, String helixParticipantId, String helixInstanceTag) { String containerProcessName = GobblinTemporalYarnTaskRunner.class.getSimpleName(); StringBuilder containerCommand = new StringBuilder() .append(ApplicationConstants.Environment.JAVA_HOME.$()).append("/bin/java") .append(" -Xmx").append((int) (container.getResource().getMemory() * this.jvmMemoryXmxRatio) - this.jvmMemoryOverheadMbs).append("M") .append(" -D").append(GobblinYarnConfigurationKeys.JVM_USER_TIMEZONE_CONFIG).append("=").append(this.containerTimezone) .append(" -D").append(GobblinYarnConfigurationKeys.GOBBLIN_YARN_CONTAINER_LOG_DIR_NAME).append("=").append(ApplicationConstants.LOG_DIR_EXPANSION_VAR) .append(" -D").append(GobblinYarnConfigurationKeys.GOBBLIN_YARN_CONTAINER_LOG_FILE_NAME).append("=").append(containerProcessName).append(".").append(ApplicationConstants.STDOUT) .append(" ").append(JvmUtils.formatJvmArguments(this.containerJvmArgs)) .append(" ").append(GobblinTemporalYarnTaskRunner.class.getName()) .append(" --").append(GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME) .append(" ").append(this.applicationName) .append(" --").append(GobblinClusterConfigurationKeys.APPLICATION_ID_OPTION_NAME) .append(" ").append(this.applicationId) .append(" --").append(GobblinClusterConfigurationKeys.HELIX_INSTANCE_NAME_OPTION_NAME) .append(" ").append(helixParticipantId); if (!Strings.isNullOrEmpty(helixInstanceTag)) { containerCommand.append(" --").append(GobblinClusterConfigurationKeys.HELIX_INSTANCE_TAGS_OPTION_NAME) .append(" ").append(helixInstanceTag); } LOGGER.info("Building " + containerProcessName); return containerCommand.append(" 1>").append(ApplicationConstants.LOG_DIR_EXPANSION_VAR).append(File.separator).append( containerProcessName).append(".").append(ApplicationConstants.STDOUT) .append(" 2>").append(ApplicationConstants.LOG_DIR_EXPANSION_VAR).append(File.separator).append( containerProcessName).append(".").append(ApplicationConstants.STDERR).toString(); } /** * Check the exit status of a completed container and see if the replacement container * should try to be started on the same node. Some exit status indicates a disk or * node failure and in such cases the replacement container should try to be started on * a different node. */ private boolean shouldStickToTheSameNode(int containerExitStatus) { switch (containerExitStatus) { case ContainerExitStatus.DISKS_FAILED: return false; case ContainerExitStatus.ABORTED: // Mostly likely this exit status is due to node failures because the // application itself will not release containers. return false; default: // Stick to the same node for other cases if host affinity is enabled. return this.containerHostAffinityEnabled; } } /** * Handle the completion of a container. A new container will be requested to replace the one * that just exited. Depending on the exit status and if container host affinity is enabled, * the new container may or may not try to be started on the same node. * <p> * A container completes in either of the following conditions: 1) some error happens in the * container and caused the container to exit, 2) the container gets killed due to some reason, * for example, if it runs over the allowed amount of virtual or physical memory, 3) the gets * preempted by the ResourceManager, or 4) the container gets stopped by the ApplicationMaster. * A replacement container is needed in all but the last case. */ protected void handleContainerCompletion(ContainerStatus containerStatus) { ContainerInfo completedContainerInfo = this.containerMap.remove(containerStatus.getContainerId()); //Get the Helix instance name for the completed container. Because callbacks are processed asynchronously, we might //encounter situations where handleContainerCompletion() is called before onContainersAllocated(), resulting in the //containerId missing from the containersMap. // We use removedContainerID to remember these containers and remove them from containerMap later when we call requestTargetNumberOfContainers method if (completedContainerInfo == null) { removedContainerID.putIfAbsent(containerStatus.getContainerId(), ""); } String completedInstanceName = UNKNOWN_HELIX_INSTANCE; String helixTag = completedContainerInfo == null ? helixInstanceTags : completedContainerInfo.getHelixTag(); if (completedContainerInfo != null) { allocatedContainerCountMap.get(helixTag).decrementAndGet(); } LOGGER.info(String.format("Container %s running Helix instance %s with tag %s has completed with exit status %d", containerStatus.getContainerId(), completedInstanceName, helixTag, containerStatus.getExitStatus())); if (!Strings.isNullOrEmpty(containerStatus.getDiagnostics())) { LOGGER.info(String.format("Received the following diagnostics information for container %s: %s", containerStatus.getContainerId(), containerStatus.getDiagnostics())); } switch(containerStatus.getExitStatus()) { case(ContainerExitStatus.ABORTED): if (handleAbortedContainer(containerStatus, completedContainerInfo, completedInstanceName)) { return; } break; case(1): // Same as linux exit status 1 Often occurs when launch_container.sh failed LOGGER.info("Exit status 1. CompletedContainerInfo={}", completedContainerInfo); break; default: break; } if (this.shutdownInProgress) { return; } if(completedContainerInfo != null) { this.helixInstanceRetryCount.putIfAbsent(completedInstanceName, new AtomicInteger(0)); int retryCount = this.helixInstanceRetryCount.get(completedInstanceName).incrementAndGet(); // Populate event metadata Optional<ImmutableMap.Builder<String, String>> eventMetadataBuilder = Optional.absent(); if (this.eventSubmitter.isPresent()) { eventMetadataBuilder = Optional.of(buildContainerStatusEventMetadata(containerStatus)); eventMetadataBuilder.get().put(GobblinYarnEventConstants.EventMetadata.HELIX_INSTANCE_ID, completedInstanceName); eventMetadataBuilder.get().put(GobblinYarnEventConstants.EventMetadata.CONTAINER_STATUS_RETRY_ATTEMPT, retryCount + ""); } if (this.helixInstanceMaxRetries > 0 && retryCount > this.helixInstanceMaxRetries) { if (this.eventSubmitter.isPresent()) { this.eventSubmitter.get() .submit(GobblinYarnEventConstants.EventNames.HELIX_INSTANCE_COMPLETION, eventMetadataBuilder.get().build()); } LOGGER.warn("Maximum number of retries has been achieved for Helix instance " + completedInstanceName); return; } // Add the Helix instance name of the completed container to the set of unused // instance names so they can be reused by a replacement container. LOGGER.info("Adding instance {} to the pool of unused instances", completedInstanceName); this.unusedHelixInstanceNames.add(completedInstanceName); if (this.eventSubmitter.isPresent()) { this.eventSubmitter.get() .submit(GobblinYarnEventConstants.EventNames.HELIX_INSTANCE_COMPLETION, eventMetadataBuilder.get().build()); } } Optional<Resource> newContainerResource = completedContainerInfo != null ? Optional.of(completedContainerInfo.getContainer().getResource()) : Optional.absent(); LOGGER.info("Requesting a new container to replace {} to run Helix instance {} with helix tag {} and resource {}", containerStatus.getContainerId(), completedInstanceName, helixTag, newContainerResource.orNull()); this.eventBus.post(new NewContainerRequest( shouldStickToTheSameNode(containerStatus.getExitStatus()) && completedContainerInfo != null ? Optional.of(completedContainerInfo.getContainer()) : Optional.absent(), newContainerResource)); } private boolean handleAbortedContainer(ContainerStatus containerStatus, ContainerInfo completedContainerInfo, String completedInstanceName) { if (this.releasedContainerCache.getIfPresent(containerStatus.getContainerId()) != null) { LOGGER.info("Container release requested, so not spawning a replacement for containerId {}", containerStatus.getContainerId()); if (completedContainerInfo != null) { LOGGER.info("Adding instance {} to the pool of unused instances", completedInstanceName); this.unusedHelixInstanceNames.add(completedInstanceName); } return true; } LOGGER.info("Container {} aborted due to lost NM", containerStatus.getContainerId()); return false; } private ImmutableMap.Builder<String, String> buildContainerStatusEventMetadata(ContainerStatus containerStatus) { ImmutableMap.Builder<String, String> eventMetadataBuilder = new ImmutableMap.Builder<>(); eventMetadataBuilder.put(GobblinYarnMetricTagNames.CONTAINER_ID, containerStatus.getContainerId().toString()); eventMetadataBuilder.put(GobblinYarnEventConstants.EventMetadata.CONTAINER_STATUS_CONTAINER_STATE, containerStatus.getState().toString()); if (ContainerExitStatus.INVALID != containerStatus.getExitStatus()) { eventMetadataBuilder.put(GobblinYarnEventConstants.EventMetadata.CONTAINER_STATUS_EXIT_STATUS, containerStatus.getExitStatus() + ""); } if (!Strings.isNullOrEmpty(containerStatus.getDiagnostics())) { eventMetadataBuilder.put(GobblinYarnEventConstants.EventMetadata.CONTAINER_STATUS_EXIT_DIAGNOSTICS, containerStatus.getDiagnostics()); } return eventMetadataBuilder; } /** * A custom implementation of {@link AMRMClientAsync.CallbackHandler}. */ private class AMRMClientCallbackHandler implements AMRMClientAsync.CallbackHandler { private volatile boolean done = false; @Override public void onContainersCompleted(List<ContainerStatus> statuses) { for (ContainerStatus containerStatus : statuses) { handleContainerCompletion(containerStatus); } } @Override public void onContainersAllocated(List<Container> containers) { for (final Container container : containers) { String containerId = container.getId().toString(); String containerHelixTag = helixInstanceTags; if (eventSubmitter.isPresent()) { eventSubmitter.get().submit(GobblinYarnEventConstants.EventNames.CONTAINER_ALLOCATION, GobblinYarnMetricTagNames.CONTAINER_ID, containerId); } LOGGER.info("Container {} has been allocated with resource {} for helix tag {}", container.getId(), container.getResource(), containerHelixTag); //Iterate over the (thread-safe) set of unused instances to find the first instance that is not currently live. //Once we find a candidate instance, it is removed from the set. String instanceName = null; //Ensure that updates to unusedHelixInstanceNames are visible to other threads that might concurrently //invoke the callback on container allocation. synchronized (this) { Iterator<String> iterator = unusedHelixInstanceNames.iterator(); while (iterator.hasNext()) { instanceName = iterator.next(); } } ContainerInfo containerInfo = new ContainerInfo(container, instanceName, containerHelixTag); containerMap.put(container.getId(), containerInfo); allocatedContainerCountMap.putIfAbsent(containerHelixTag, new AtomicInteger(0)); allocatedContainerCountMap.get(containerHelixTag).incrementAndGet(); // Find matching requests and remove the request (YARN-660). We the scheduler are responsible // for cleaning up requests after allocation based on the design in the described ticket. // YARN does not have a delta request API and the requests are not cleaned up automatically. // Try finding a match first with the host as the resource name then fall back to any resource match. // Also see YARN-1902. Container count will explode without this logic for removing container requests. List<? extends Collection<AMRMClient.ContainerRequest>> matchingRequests = amrmClientAsync .getMatchingRequests(container.getPriority(), container.getNodeHttpAddress(), container.getResource()); if (matchingRequests.isEmpty()) { LOGGER.debug("Matching request by host {} not found", container.getNodeHttpAddress()); matchingRequests = amrmClientAsync .getMatchingRequests(container.getPriority(), ResourceRequest.ANY, container.getResource()); } if (!matchingRequests.isEmpty()) { AMRMClient.ContainerRequest firstMatchingContainerRequest = matchingRequests.get(0).iterator().next(); LOGGER.debug("Found matching requests {}, removing first matching request {}", matchingRequests, firstMatchingContainerRequest); amrmClientAsync.removeContainerRequest(firstMatchingContainerRequest); } containerLaunchExecutor.submit(new Runnable() { @Override public void run() { try { LOGGER.info("Starting container " + containerId); nmClientAsync.startContainerAsync(container, newContainerLaunchContext(containerInfo)); } catch (IOException ioe) { LOGGER.error("Failed to start container " + containerId, ioe); } } }); } } @Override public void onShutdownRequest() { if (eventSubmitter.isPresent()) { eventSubmitter.get().submit(GobblinYarnEventConstants.EventNames.SHUTDOWN_REQUEST); } LOGGER.info("Received shutdown request from the ResourceManager"); this.done = true; eventBus.post(new ClusterManagerShutdownRequest()); } @Override public void onNodesUpdated(List<NodeReport> updatedNodes) { for (NodeReport nodeReport : updatedNodes) { LOGGER.info("Received node update report: " + nodeReport); } } @Override public float getProgress() { return this.done ? 1.0f : 0.0f; } @Override public void onError(Throwable t) { if (eventSubmitter.isPresent()) { eventSubmitter.get().submit(GobblinYarnEventConstants.EventNames.ERROR, GobblinYarnEventConstants.EventMetadata.ERROR_EXCEPTION, Throwables.getStackTraceAsString(t)); } LOGGER.error("Received error: " + t, t); this.done = true; eventBus.post(new ClusterManagerShutdownRequest()); } } /** * A custom implementation of {@link NMClientAsync.CallbackHandler}. */ class NMClientCallbackHandler implements NMClientAsync.CallbackHandler { @Override public void onContainerStarted(ContainerId containerId, Map<String, ByteBuffer> allServiceResponse) { if (eventSubmitter.isPresent()) { eventSubmitter.get().submit(GobblinYarnEventConstants.EventNames.CONTAINER_STARTED, GobblinYarnMetricTagNames.CONTAINER_ID, containerId.toString()); } LOGGER.info(String.format("Container %s has been started", containerId)); } @Override public void onContainerStatusReceived(ContainerId containerId, ContainerStatus containerStatus) { if (eventSubmitter.isPresent()) { eventSubmitter.get().submit(GobblinYarnEventConstants.EventNames.CONTAINER_STATUS_RECEIVED, buildContainerStatusEventMetadata(containerStatus).build()); } LOGGER.info(String.format("Received container status for container %s: %s", containerId, containerStatus)); } @Override public void onContainerStopped(ContainerId containerId) { if (eventSubmitter.isPresent()) { eventSubmitter.get().submit(GobblinYarnEventConstants.EventNames.CONTAINER_STOPPED, GobblinYarnMetricTagNames.CONTAINER_ID, containerId.toString()); } LOGGER.info(String.format("Container %s has been stopped", containerId)); if (containerMap.isEmpty()) { synchronized (allContainersStopped) { allContainersStopped.notify(); } } } @Override public void onStartContainerError(ContainerId containerId, Throwable t) { if (eventSubmitter.isPresent()) { eventSubmitter.get().submit(GobblinYarnEventConstants.EventNames.CONTAINER_START_ERROR, GobblinYarnMetricTagNames.CONTAINER_ID, containerId.toString(), GobblinYarnEventConstants.EventMetadata.ERROR_EXCEPTION, Throwables.getStackTraceAsString(t)); } LOGGER.error(String.format("Failed to start container %s due to error %s", containerId, t)); } @Override public void onGetContainerStatusError(ContainerId containerId, Throwable t) { if (eventSubmitter.isPresent()) { eventSubmitter.get().submit(GobblinYarnEventConstants.EventNames.CONTAINER_GET_STATUS_ERROR, GobblinYarnMetricTagNames.CONTAINER_ID, containerId.toString(), GobblinYarnEventConstants.EventMetadata.ERROR_EXCEPTION, Throwables.getStackTraceAsString(t)); } LOGGER.error(String.format("Failed to get status for container %s due to error %s", containerId, t)); } @Override public void onStopContainerError(ContainerId containerId, Throwable t) { if (eventSubmitter.isPresent()) { eventSubmitter.get().submit(GobblinYarnEventConstants.EventNames.CONTAINER_STOP_ERROR, GobblinYarnMetricTagNames.CONTAINER_ID, containerId.toString(), GobblinYarnEventConstants.EventMetadata.ERROR_EXCEPTION, Throwables.getStackTraceAsString(t)); } LOGGER.error(String.format("Failed to stop container %s due to error %s", containerId, t)); } } // Class encapsulates Container instances, Helix participant IDs of the containers, Helix Tag, and // initial startup command @Getter class ContainerInfo { private final Container container; private final String helixParticipantId; private final String helixTag; private final String startupCommand; public ContainerInfo(Container container, String helixParticipantId, String helixTag) { this.container = container; this.helixParticipantId = helixParticipantId; this.helixTag = helixTag; this.startupCommand = YarnService.this.buildContainerCommand(container, helixParticipantId, helixTag); } @Override public String toString() { return String.format("ContainerInfo{ container=%s, helixParticipantId=%s, helixTag=%s, startupCommand=%s }", container.getId(), helixParticipantId, helixTag, startupCommand); } } }
3,976
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/yarn/GobblinTemporalApplicationMaster.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.yarn; import com.google.common.base.Optional; import com.google.common.util.concurrent.Service; import com.typesafe.config.Config; import com.typesafe.config.ConfigFactory; import com.typesafe.config.ConfigValueFactory; import java.util.List; import lombok.Getter; import org.slf4j.Logger; import org.slf4j.LoggerFactory; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.cluster.GobblinClusterConfigurationKeys; import org.apache.gobblin.cluster.GobblinClusterUtils; import org.apache.gobblin.temporal.cluster.GobblinTemporalClusterManager; import org.apache.gobblin.util.ConfigUtils; import org.apache.gobblin.util.JvmUtils; import org.apache.gobblin.util.PathUtils; import org.apache.gobblin.util.logs.Log4jConfigurationHelper; import org.apache.gobblin.util.logs.LogCopier; import org.apache.gobblin.util.reflection.GobblinConstructorUtils; import org.apache.gobblin.yarn.GobblinYarnConfigurationKeys; import org.apache.gobblin.yarn.GobblinYarnLogSource; import org.apache.gobblin.yarn.YarnContainerSecurityManager; import org.apache.gobblin.yarn.YarnHelixUtils; import org.apache.commons.cli.CommandLine; import org.apache.commons.cli.DefaultParser; import org.apache.commons.cli.HelpFormatter; import org.apache.commons.cli.Options; import org.apache.commons.cli.ParseException; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.apache.hadoop.security.UserGroupInformation; import org.apache.hadoop.yarn.api.ApplicationConstants; import org.apache.hadoop.yarn.api.records.ContainerId; import org.apache.hadoop.yarn.conf.YarnConfiguration; import org.apache.hadoop.yarn.util.ConverterUtils; /** * The Yarn ApplicationMaster class for Gobblin using Temporal. * * <p> * This class runs the {@link YarnService} for all Yarn-related stuffs like ApplicationMaster registration * and un-registration and Yarn container provisioning. * </p> * */ @Alpha public class GobblinTemporalApplicationMaster extends GobblinTemporalClusterManager { private static final Logger LOGGER = LoggerFactory.getLogger(GobblinTemporalApplicationMaster.class); @Getter private final YarnService _yarnService; private LogCopier logCopier; public GobblinTemporalApplicationMaster(String applicationName, String applicationId, ContainerId containerId, Config config, YarnConfiguration yarnConfiguration) throws Exception { super(applicationName, applicationId, config.withValue(GobblinYarnConfigurationKeys.CONTAINER_NUM_KEY, ConfigValueFactory.fromAnyRef(YarnHelixUtils.getContainerNum(containerId.toString()))), Optional.<Path>absent()); String containerLogDir = config.getString(GobblinYarnConfigurationKeys.LOGS_SINK_ROOT_DIR_KEY); GobblinYarnLogSource gobblinYarnLogSource = new GobblinYarnLogSource(); if (gobblinYarnLogSource.isLogSourcePresent()) { Path appWorkDir = PathUtils.combinePaths(containerLogDir, GobblinClusterUtils.getAppWorkDirPath(this.clusterName, this.applicationId), "AppMaster"); logCopier = gobblinYarnLogSource.buildLogCopier(this.config, containerId.toString(), this.fs, appWorkDir); this.applicationLauncher .addService(logCopier); } YarnHelixUtils.setYarnClassPath(config, yarnConfiguration); YarnHelixUtils.setAdditionalYarnClassPath(config, yarnConfiguration); this._yarnService = buildTemporalYarnService(this.config, applicationName, this.applicationId, yarnConfiguration, this.fs); this.applicationLauncher.addService(this._yarnService); if (UserGroupInformation.isSecurityEnabled()) { LOGGER.info("Adding YarnContainerSecurityManager since security is enabled"); this.applicationLauncher.addService(buildYarnContainerSecurityManager(this.config, this.fs)); } // Add additional services List<String> serviceClassNames = ConfigUtils.getStringList(this.config, GobblinYarnConfigurationKeys.APP_MASTER_SERVICE_CLASSES); for (String serviceClassName : serviceClassNames) { Class<?> serviceClass = Class.forName(serviceClassName); this.applicationLauncher.addService((Service) GobblinConstructorUtils.invokeLongestConstructor(serviceClass, this)); } } /** * Build the {@link YarnService} for the Application Master. */ protected YarnService buildTemporalYarnService(Config config, String applicationName, String applicationId, YarnConfiguration yarnConfiguration, FileSystem fs) throws Exception { return new YarnService(config, applicationName, applicationId, yarnConfiguration, fs, this.eventBus); } /** * Build the {@link YarnTemporalAppMasterSecurityManager} for the Application Master. */ private YarnContainerSecurityManager buildYarnContainerSecurityManager(Config config, FileSystem fs) { return new YarnTemporalAppMasterSecurityManager(config, fs, this.eventBus, this.logCopier, this._yarnService); } private static Options buildOptions() { Options options = new Options(); options.addOption("a", GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME, true, "Yarn application name"); options.addOption("d", GobblinClusterConfigurationKeys.APPLICATION_ID_OPTION_NAME, true, "Yarn application id"); return options; } private static void printUsage(Options options) { HelpFormatter formatter = new HelpFormatter(); formatter.printHelp(GobblinTemporalApplicationMaster.class.getSimpleName(), options); } public static void main(String[] args) throws Exception { Options options = buildOptions(); try { CommandLine cmd = new DefaultParser().parse(options, args); if (!cmd.hasOption(GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME) || (!cmd.hasOption(GobblinClusterConfigurationKeys.APPLICATION_ID_OPTION_NAME))) { printUsage(options); System.exit(1); } //Because AM is restarted with the original AppSubmissionContext, it may have outdated delegation tokens. //So the refreshed tokens should be added into the container's UGI before any HDFS/Hive/RM access is performed. YarnHelixUtils.updateToken(GobblinYarnConfigurationKeys.TOKEN_FILE_NAME); Log4jConfigurationHelper.updateLog4jConfiguration(GobblinTemporalApplicationMaster.class, GobblinYarnConfigurationKeys.GOBBLIN_YARN_LOG4J_CONFIGURATION_FILE, GobblinYarnConfigurationKeys.GOBBLIN_YARN_LOG4J_CONFIGURATION_FILE); LOGGER.info(JvmUtils.getJvmInputArguments()); ContainerId containerId = ConverterUtils.toContainerId(System.getenv().get(ApplicationConstants.Environment.CONTAINER_ID.key())); try (GobblinTemporalApplicationMaster applicationMaster = new GobblinTemporalApplicationMaster( cmd.getOptionValue(GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME), cmd.getOptionValue(GobblinClusterConfigurationKeys.APPLICATION_ID_OPTION_NAME), containerId, ConfigFactory.load(), new YarnConfiguration())) { applicationMaster.start(); } } catch (ParseException pe) { printUsage(options); System.exit(1); } } }
3,977
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/yarn/GobblinTemporalYarnTaskRunner.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.yarn; import java.util.ArrayList; import java.util.List; import org.apache.commons.cli.CommandLine; import org.apache.commons.cli.DefaultParser; import org.apache.commons.cli.Options; import org.apache.commons.cli.ParseException; import org.apache.gobblin.yarn.GobblinYarnConfigurationKeys; import org.apache.gobblin.yarn.GobblinYarnLogSource; import org.apache.gobblin.yarn.YarnContainerSecurityManager; import org.apache.gobblin.yarn.YarnHelixUtils; import org.apache.hadoop.fs.Path; import org.apache.hadoop.security.UserGroupInformation; import org.apache.hadoop.yarn.api.ApplicationConstants; import org.apache.hadoop.yarn.api.records.ContainerId; import org.apache.hadoop.yarn.util.ConverterUtils; import org.slf4j.Logger; import org.slf4j.LoggerFactory; import com.google.common.base.Optional; import com.google.common.util.concurrent.Service; import com.typesafe.config.Config; import com.typesafe.config.ConfigFactory; import com.typesafe.config.ConfigValueFactory; import org.apache.gobblin.cluster.GobblinClusterConfigurationKeys; import org.apache.gobblin.cluster.GobblinClusterUtils; import org.apache.gobblin.cluster.GobblinTaskRunner; import org.apache.gobblin.temporal.cluster.GobblinTemporalTaskRunner; import org.apache.gobblin.util.JvmUtils; import org.apache.gobblin.util.logs.Log4jConfigurationHelper; import org.apache.gobblin.util.logs.LogCopier; public class GobblinTemporalYarnTaskRunner extends GobblinTemporalTaskRunner { private static final Logger LOGGER = LoggerFactory.getLogger(GobblinTaskRunner.class); public GobblinTemporalYarnTaskRunner(String applicationName, String applicationId, ContainerId containerId, Config config, Optional<Path> appWorkDirOptional) throws Exception { super(applicationName, applicationId, getTaskRunnerId(containerId), config .withValue(GobblinYarnConfigurationKeys.CONTAINER_NUM_KEY, ConfigValueFactory.fromAnyRef(YarnHelixUtils.getContainerNum(containerId.toString()))), appWorkDirOptional); } @Override public List<Service> getServices() { List<Service> services = new ArrayList<>(); services.addAll(super.getServices()); LogCopier logCopier = null; if (clusterConfig.hasPath(GobblinYarnConfigurationKeys.LOGS_SINK_ROOT_DIR_KEY)) { GobblinYarnLogSource gobblinYarnLogSource = new GobblinYarnLogSource(); String containerLogDir = clusterConfig.getString(GobblinYarnConfigurationKeys.LOGS_SINK_ROOT_DIR_KEY); if (gobblinYarnLogSource.isLogSourcePresent()) { try { logCopier = gobblinYarnLogSource.buildLogCopier(this.clusterConfig, this.taskRunnerId, this.fs, new Path(containerLogDir, GobblinClusterUtils.getAppWorkDirPath(this.applicationName, this.applicationId))); services.add(logCopier); } catch (Exception e) { LOGGER.warn("Cannot add LogCopier service to the service manager due to", e); } } } if (UserGroupInformation.isSecurityEnabled()) { LOGGER.info("Adding YarnContainerSecurityManager since security is enabled"); services.add(new YarnContainerSecurityManager(this.clusterConfig, this.fs, this.eventBus, logCopier)); } return services; } private static String getApplicationId(ContainerId containerId) { return containerId.getApplicationAttemptId().getApplicationId().toString(); } private static String getTaskRunnerId(ContainerId containerId) { return containerId.toString(); } public static void main(String[] args) { LOGGER.info("Starting GobblinTemporalYarnTaskRunner"); Options options = GobblinTemporalTaskRunner.buildOptions(); try { CommandLine cmd = new DefaultParser().parse(options, args); if (!cmd.hasOption(GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME) || !cmd .hasOption(GobblinClusterConfigurationKeys.APPLICATION_ID_OPTION_NAME)) { GobblinTemporalTaskRunner.printUsage(options); System.exit(1); } Log4jConfigurationHelper.updateLog4jConfiguration(GobblinTaskRunner.class, GobblinYarnConfigurationKeys.GOBBLIN_YARN_LOG4J_CONFIGURATION_FILE, GobblinYarnConfigurationKeys.GOBBLIN_YARN_LOG4J_CONFIGURATION_FILE); LOGGER.info(JvmUtils.getJvmInputArguments()); ContainerId containerId = ConverterUtils.toContainerId(System.getenv().get(ApplicationConstants.Environment.CONTAINER_ID.key())); String applicationName = cmd.getOptionValue(GobblinClusterConfigurationKeys.APPLICATION_NAME_OPTION_NAME); String applicationId = cmd.getOptionValue(GobblinClusterConfigurationKeys.APPLICATION_ID_OPTION_NAME); Config config = ConfigFactory.load(); GobblinTemporalTaskRunner gobblinTemporalTaskRunner = new GobblinTemporalYarnTaskRunner(applicationName, applicationId, containerId, config, Optional.<Path>absent()); gobblinTemporalTaskRunner.start(); } catch (ParseException pe) { GobblinTemporalTaskRunner.printUsage(options); System.exit(1); } catch (Throwable t) { // Ideally, we should not be catching non-recoverable exceptions and errors. However, // simply propagating the exception may prevent the container exit due to the presence of non-daemon threads present // in the application. Hence, we catch this exception to invoke System.exit() which in turn ensures that all non-daemon threads are killed. LOGGER.error("Exception encountered: {}", t); System.exit(1); } } }
3,978
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen/launcher/GenArbitraryLoadJobLauncher.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.loadgen.launcher; import java.util.List; import java.util.Optional; import java.util.Properties; import java.util.concurrent.ConcurrentHashMap; import lombok.extern.slf4j.Slf4j; import io.temporal.client.WorkflowOptions; import org.apache.hadoop.fs.Path; import org.apache.gobblin.annotation.Alpha; import org.apache.gobblin.metrics.Tag; import org.apache.gobblin.runtime.JobLauncher; import org.apache.gobblin.source.workunit.WorkUnit; import org.apache.gobblin.temporal.cluster.GobblinTemporalTaskRunner; import org.apache.gobblin.temporal.joblauncher.GobblinTemporalJobLauncher; import org.apache.gobblin.temporal.joblauncher.GobblinTemporalJobScheduler; import org.apache.gobblin.temporal.loadgen.work.IllustrationItem; import org.apache.gobblin.temporal.loadgen.work.SimpleGeneratedWorkload; import org.apache.gobblin.temporal.util.nesting.work.WorkflowAddr; import org.apache.gobblin.temporal.util.nesting.work.Workload; import org.apache.gobblin.temporal.util.nesting.workflow.NestingExecWorkflow; import org.apache.gobblin.util.PropertiesUtils; import static org.apache.gobblin.temporal.GobblinTemporalConfigurationKeys.GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_PREFIX; /** * A {@link JobLauncher} for the initial triggering of a Temporal workflow that generates arbitrary load of many * activities nested beneath a single subsuming super-workflow. see: {@link NestingExecWorkflow} * * <p> * This class is instantiated by the {@link GobblinTemporalJobScheduler#buildJobLauncher(Properties)} on every job submission to launch the Gobblin job. * The actual task execution happens in the {@link GobblinTemporalTaskRunner}, usually in a different process. * </p> */ @Alpha @Slf4j public class GenArbitraryLoadJobLauncher extends GobblinTemporalJobLauncher { public static final String GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_NUM_ACTIVITIES = GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_PREFIX + "num.activities"; public static final String GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_MAX_BRANCHES_PER_TREE = GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_PREFIX + "max.branches.per.tree"; public static final String GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_MAX_SUB_TREES_PER_TREE = GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_PREFIX + "max.sub.trees.per.tree"; public GenArbitraryLoadJobLauncher( Properties jobProps, Path appWorkDir, List<? extends Tag<?>> metadataTags, ConcurrentHashMap<String, Boolean> runningMap ) throws Exception { super(jobProps, appWorkDir, metadataTags, runningMap); } @Override public void submitJob(List<WorkUnit> workunits) { int numActivities = PropertiesUtils.getRequiredPropAsInt(this.jobProps, GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_NUM_ACTIVITIES); int maxBranchesPerTree = PropertiesUtils.getRequiredPropAsInt(this.jobProps, GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_MAX_BRANCHES_PER_TREE); int maxSubTreesPerTree = PropertiesUtils.getRequiredPropAsInt(this.jobProps, GOBBLIN_TEMPORAL_JOB_LAUNCHER_ARG_MAX_SUB_TREES_PER_TREE); Workload<IllustrationItem> workload = SimpleGeneratedWorkload.createAs(numActivities); WorkflowOptions options = WorkflowOptions.newBuilder().setTaskQueue(this.queueName).build(); // WARNING: although type param must agree w/ that of `workload`, it's entirely unverified by type checker! // ...and more to the point, mismatch would occur at runtime (`performWorkload` on the workflow type given to the stub)! NestingExecWorkflow<IllustrationItem> workflow = this.client.newWorkflowStub(NestingExecWorkflow.class, options); workflow.performWorkload(WorkflowAddr.ROOT, workload, 0, maxBranchesPerTree, maxSubTreesPerTree, Optional.empty()); } }
3,979
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen/activity/IllustrationItemActivity.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.loadgen.activity; import io.temporal.activity.ActivityInterface; import io.temporal.activity.ActivityMethod; import org.apache.gobblin.temporal.loadgen.work.IllustrationItem; /** * Activity for processing {@link IllustrationItem}s * * CAUTION/FINDING: an `@ActivityInterface` must not be parameterized (e.g. here, by WORK_ITEM), as doing so results in: * io.temporal.failure.ApplicationFailure: message='class java.util.LinkedHashMap cannot be cast to class * org.apache.gobblin.temporal.loadgen.work.IllustrationItem', type='java.lang.ClassCastException' */ @ActivityInterface public interface IllustrationItemActivity { @ActivityMethod String handleItem(IllustrationItem item); }
3,980
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen/activity
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen/activity/impl/IllustrationItemActivityImpl.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.loadgen.activity.impl; import lombok.extern.slf4j.Slf4j; import org.apache.gobblin.temporal.loadgen.activity.IllustrationItemActivity; import org.apache.gobblin.temporal.loadgen.work.IllustrationItem; @Slf4j public class IllustrationItemActivityImpl implements IllustrationItemActivity { @Override public String handleItem(final IllustrationItem item) { log.info("Now illustrating - '" + item.getName() + "'"); return item.getName(); } }
3,981
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen/work/IllustrationItem.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.loadgen.work; import lombok.Data; import lombok.NoArgsConstructor; import lombok.NonNull; import lombok.RequiredArgsConstructor; /** Generally, this would specify what "work" needs performing plus how to perform, but for now merely a unique name (to log) */ @Data @NoArgsConstructor // IMPORTANT: for jackson (de)serialization @RequiredArgsConstructor public class IllustrationItem { @NonNull private String name; }
3,982
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen/work/SimpleGeneratedWorkload.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.loadgen.work; import com.fasterxml.jackson.annotation.JsonIgnore; import java.util.List; import java.util.Optional; import java.util.stream.Collectors; import java.util.stream.IntStream; import lombok.AccessLevel; import org.apache.gobblin.temporal.util.nesting.work.SeqBackedWorkSpan; import org.apache.gobblin.temporal.util.nesting.work.Workload; /** Example, illustration workload that synthesizes its work items; genuine {@link Workload}s generally arise from query/calc */ @lombok.AllArgsConstructor(access = AccessLevel.PRIVATE) @lombok.NoArgsConstructor // IMPORTANT: for jackson (de)serialization @lombok.ToString public class SimpleGeneratedWorkload implements Workload<IllustrationItem> { private int numItems; /** Factory method */ public static SimpleGeneratedWorkload createAs(final int numItems) { return new SimpleGeneratedWorkload(numItems); } @Override public Optional<Workload.WorkSpan<IllustrationItem>> getSpan(final int startIndex, final int numElements) { if (startIndex >= numItems || startIndex < 0) { return Optional.empty(); } else { List<IllustrationItem> elems = IntStream.range(startIndex, Math.min(startIndex + numElements, numItems)) .mapToObj(n -> new IllustrationItem("item-" + n + "-of-" + numItems)) .collect(Collectors.toList()); return Optional.of(new SeqBackedWorkSpan<>(elems, startIndex)); } } @Override public boolean isIndexKnownToExceed(final int index) { return isDefiniteSize() && index >= numItems; } @Override @JsonIgnore // (because no-arg method resembles 'java bean property') public boolean isDefiniteSize() { return true; } }
3,983
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen/workflow
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen/workflow/impl/NestingExecOfIllustrationItemActivityWorkflowImpl.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.loadgen.workflow.impl; import io.temporal.activity.ActivityOptions; import io.temporal.common.RetryOptions; import io.temporal.workflow.Async; import io.temporal.workflow.Promise; import io.temporal.workflow.Workflow; import java.time.Duration; import org.apache.gobblin.temporal.loadgen.activity.IllustrationItemActivity; import org.apache.gobblin.temporal.loadgen.work.IllustrationItem; import org.apache.gobblin.temporal.util.nesting.workflow.AbstractNestingExecWorkflowImpl; /** {@link org.apache.gobblin.temporal.util.nesting.workflow.NestingExecWorkflow} for {@link IllustrationItem} */ public class NestingExecOfIllustrationItemActivityWorkflowImpl extends AbstractNestingExecWorkflowImpl<IllustrationItem, String> { // RetryOptions specify how to automatically handle retries when Activities fail. private static final RetryOptions ACTIVITY_RETRY_OPTS = RetryOptions.newBuilder() .setInitialInterval(Duration.ofSeconds(1)) .setMaximumInterval(Duration.ofSeconds(100)) .setBackoffCoefficient(2) .setMaximumAttempts(3) .build(); private static final ActivityOptions ACTIVITY_OPTS = ActivityOptions.newBuilder() .setStartToCloseTimeout(Duration.ofSeconds(10)) .setRetryOptions(ACTIVITY_RETRY_OPTS) .build(); private final IllustrationItemActivity activityStub = Workflow.newActivityStub(IllustrationItemActivity.class, ACTIVITY_OPTS); @Override protected Promise<String> launchAsyncActivity(final IllustrationItem item) { return Async.function(activityStub::handleItem, item); } }
3,984
0
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen
Create_ds/gobblin/gobblin-temporal/src/main/java/org/apache/gobblin/temporal/loadgen/worker/ArbitraryLoadWorker.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.temporal.loadgen.worker; import com.typesafe.config.Config; import io.temporal.client.WorkflowClient; import org.apache.gobblin.temporal.cluster.AbstractTemporalWorker; import org.apache.gobblin.temporal.loadgen.activity.impl.IllustrationItemActivityImpl; import org.apache.gobblin.temporal.loadgen.workflow.impl.NestingExecOfIllustrationItemActivityWorkflowImpl; /** Worker for {@link NestingExecOfIllustrationItemActivityWorkflowImpl} and said activity impl */ public class ArbitraryLoadWorker extends AbstractTemporalWorker { public ArbitraryLoadWorker(Config config, WorkflowClient workflowClient) { super(config, workflowClient); } @Override protected Class<?>[] getWorkflowImplClasses() { return new Class[] { NestingExecOfIllustrationItemActivityWorkflowImpl.class }; } @Override protected Object[] getActivityImplInstances() { return new Object[] { new IllustrationItemActivityImpl() }; } }
3,985
0
Create_ds/gobblin/gobblin-audit/src/test/java/org/apache/gobblin/audit
Create_ds/gobblin/gobblin-audit/src/test/java/org/apache/gobblin/audit/values/MockSink.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values; import org.apache.gobblin.annotation.Alias; import org.apache.gobblin.audit.values.auditor.ValueAuditRuntimeMetadata; import org.apache.gobblin.audit.values.sink.AuditSink; import java.io.IOException; import org.apache.avro.generic.GenericRecord; import com.typesafe.config.Config; @Alias(value = "MockSink") public class MockSink implements AuditSink { public MockSink(Config config, ValueAuditRuntimeMetadata auditMetadata) {} @Override public void close() throws IOException {} @Override public void write(GenericRecord record) throws IOException {} }
3,986
0
Create_ds/gobblin/gobblin-audit/src/test/java/org/apache/gobblin/audit
Create_ds/gobblin/gobblin-audit/src/test/java/org/apache/gobblin/audit/values/FsAuditSinkTest.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values; import org.apache.gobblin.audit.values.auditor.ValueAuditRuntimeMetadata; import org.apache.gobblin.audit.values.sink.FsAuditSink; import java.io.File; import org.apache.avro.Schema; import org.apache.avro.SchemaBuilder; import org.apache.avro.file.DataFileReader; import org.apache.avro.generic.GenericDatumReader; import org.apache.avro.generic.GenericRecord; import org.apache.avro.generic.GenericRecordBuilder; import org.apache.commons.lang3.RandomStringUtils; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import org.testng.Assert; import org.testng.annotations.Test; import com.typesafe.config.ConfigFactory; @Test(groups = { "gobblin.audit.values" }) public class FsAuditSinkTest { @Test public void testWrite() throws Exception { Schema testSchema = SchemaBuilder.record("test").fields().name("f1").type().stringType().noDefault().endRecord(); GenericRecord r = new GenericRecordBuilder(testSchema).set("f1", "v1").build(); ValueAuditRuntimeMetadata auditMetadata = ValueAuditRuntimeMetadata.builder("db", "tb", testSchema).snapshotId(RandomStringUtils.randomAlphanumeric(5)) .partFileName("part-1.avro").build(); File auditFile = null; Path auditDir = null; try (FsAuditSink auditSink = new FsAuditSink(ConfigFactory.empty(), auditMetadata);) { auditFile = new File(auditSink.getAuditFilePath().toString()); auditDir = auditSink.getAuditDirPath(); auditSink.write(r); } catch (Exception e){ FileSystem.get(new Configuration()).delete(auditDir, true); throw e; } GenericDatumReader<GenericRecord> reader = new GenericDatumReader<>(testSchema); try (DataFileReader<GenericRecord> dataFileReader = new DataFileReader<>(auditFile, reader);) { while (dataFileReader.hasNext()) { GenericRecord recRead = dataFileReader.next(); Assert.assertEquals(recRead, r); break; } Assert.assertEquals(dataFileReader.hasNext(), false); } finally { FileSystem.get(new Configuration()).delete(auditDir, true); } } }
3,987
0
Create_ds/gobblin/gobblin-audit/src/test/java/org/apache/gobblin/audit
Create_ds/gobblin/gobblin-audit/src/test/java/org/apache/gobblin/audit/values/ValueAuditGeneratorTest.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values; import static org.mockito.Mockito.mock; import org.apache.gobblin.audit.values.auditor.ValueAuditGenerator; import org.apache.gobblin.audit.values.auditor.ValueAuditRuntimeMetadata; import org.apache.gobblin.audit.values.policy.column.ProjectAllColumnProjectionPolicy; import org.apache.gobblin.audit.values.policy.row.SelectAllRowSelectionPolicy; import org.apache.avro.generic.GenericRecord; import org.mockito.Mockito; import org.testng.Assert; import org.testng.annotations.Test; import com.google.common.collect.ImmutableMap; import com.typesafe.config.Config; import com.typesafe.config.ConfigFactory; @Test(groups = {"gobblin.audit.values"}) public class ValueAuditGeneratorTest { @Test public void testConstructor() throws Exception { Config config =ConfigFactory.parseMap(ImmutableMap.of( "columnProjection.class", ProjectAllColumnProjectionPolicy.class.getCanonicalName(), "rowSelection.class", SelectAllRowSelectionPolicy.class.getCanonicalName(), "auditSink.class", MockSink.class.getCanonicalName())); ValueAuditRuntimeMetadata runtimeMetadata = mock(ValueAuditRuntimeMetadata.class, Mockito.RETURNS_SMART_NULLS); ValueAuditGenerator auditGenerator = ValueAuditGenerator.create(config, runtimeMetadata); auditGenerator.audit(mock(GenericRecord.class, Mockito.RETURNS_SMART_NULLS)); Assert.assertEquals(auditGenerator.getRowSelectionPolicy().getClass().getCanonicalName(), SelectAllRowSelectionPolicy.class.getCanonicalName()); Assert.assertEquals(auditGenerator.getColumnProjectionPolicy().getClass().getCanonicalName(), ProjectAllColumnProjectionPolicy.class.getCanonicalName()); Assert.assertEquals(auditGenerator.getAuditSink().getClass().getCanonicalName(), MockSink.class.getCanonicalName()); } @Test public void testConstructorWithAlias() throws Exception { Config config =ConfigFactory.parseMap(ImmutableMap.of( "columnProjection.class", "ProjectAll", "rowSelection.class", "SelectAll", "auditSink.class", "MockSink")); ValueAuditRuntimeMetadata runtimeMetadata = mock(ValueAuditRuntimeMetadata.class, Mockito.RETURNS_SMART_NULLS); ValueAuditGenerator auditGenerator = ValueAuditGenerator.create(config, runtimeMetadata); auditGenerator.audit(mock(GenericRecord.class, Mockito.RETURNS_SMART_NULLS)); Assert.assertEquals(auditGenerator.getRowSelectionPolicy().getClass().getCanonicalName(), SelectAllRowSelectionPolicy.class.getCanonicalName()); Assert.assertEquals(auditGenerator.getColumnProjectionPolicy().getClass().getCanonicalName(), ProjectAllColumnProjectionPolicy.class.getCanonicalName()); Assert.assertEquals(auditGenerator.getAuditSink().getClass().getCanonicalName(), MockSink.class.getCanonicalName()); } }
3,988
0
Create_ds/gobblin/gobblin-audit/src/test/java/org/apache/gobblin/audit
Create_ds/gobblin/gobblin-audit/src/test/java/org/apache/gobblin/audit/values/ValueAuditRuntimeMetadataTest.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values; import static org.mockito.Mockito.mock; import org.apache.gobblin.audit.values.auditor.ValueAuditRuntimeMetadata; import org.apache.gobblin.audit.values.auditor.ValueAuditRuntimeMetadata.Phase; import org.apache.avro.Schema; import org.mockito.Mockito; import org.testng.Assert; import org.testng.annotations.Test; @Test(groups = { "gobblin.audit.values" }) public class ValueAuditRuntimeMetadataTest { @Test public void testBuilderWithDefaults() throws Exception { Schema mockSchema = mock(Schema.class, Mockito.RETURNS_SMART_NULLS); ValueAuditRuntimeMetadata runtimeMetadata = ValueAuditRuntimeMetadata.builder("db", "t", mockSchema).build(); Assert.assertEquals(runtimeMetadata.getTableMetadata().getDatabase(), "db"); Assert.assertEquals(runtimeMetadata.getTableMetadata().getTable(), "t"); Assert.assertEquals(runtimeMetadata.getTableMetadata().getTableSchema(), mockSchema); Assert.assertEquals(runtimeMetadata.getCluster(), "NA"); Assert.assertEquals(runtimeMetadata.getDeltaId(), "NA"); Assert.assertEquals(runtimeMetadata.getExtractId(), "NA"); Assert.assertEquals(runtimeMetadata.getPartFileName(), "NA"); Assert.assertEquals(runtimeMetadata.getSnapshotId(), "NA"); Assert.assertEquals(runtimeMetadata.getPhase(), Phase.NA); } @Test public void testBuilder() throws Exception { Schema mockSchema = mock(Schema.class, Mockito.RETURNS_SMART_NULLS); ValueAuditRuntimeMetadata runtimeMetadata = ValueAuditRuntimeMetadata.builder("db", "t", mockSchema).cluster("c").deltaId("d").extractId("e") .partFileName("p").phase(Phase.AVRO_CONV).snapshotId("s").build(); Assert.assertEquals(runtimeMetadata.getTableMetadata().getDatabase(), "db"); Assert.assertEquals(runtimeMetadata.getTableMetadata().getTable(), "t"); Assert.assertEquals(runtimeMetadata.getTableMetadata().getTableSchema(), mockSchema); Assert.assertEquals(runtimeMetadata.getCluster(), "c"); Assert.assertEquals(runtimeMetadata.getDeltaId(), "d"); Assert.assertEquals(runtimeMetadata.getExtractId(), "e"); Assert.assertEquals(runtimeMetadata.getPartFileName(), "p"); Assert.assertEquals(runtimeMetadata.getSnapshotId(), "s"); Assert.assertEquals(runtimeMetadata.getPhase(), Phase.AVRO_CONV); } }
3,989
0
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/sink/FsAuditSink.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values.sink; import java.io.IOException; import java.io.OutputStream; import java.net.URI; import lombok.Getter; import org.apache.avro.file.DataFileWriter; import org.apache.avro.generic.GenericDatumWriter; import org.apache.avro.generic.GenericRecord; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.Path; import com.google.common.io.Closer; import com.typesafe.config.Config; import org.apache.gobblin.annotation.Alias; import org.apache.gobblin.audit.values.auditor.ValueAuditRuntimeMetadata; import org.apache.gobblin.configuration.ConfigurationKeys; import org.apache.gobblin.util.ConfigUtils; import org.apache.gobblin.util.PathUtils; /** * A Hadoop {@link FileSystem} based {@link AuditSink} that writes audit {@link GenericRecord}s to a file on {@link FileSystem}. * <ul> * <li> The {@link FileSystem} {@link URI} can be set using key {@link ConfigurationKeys#FS_URI_KEY}, {@link ConfigurationKeys#LOCAL_FS_URI} is used by default. * <li> All audit files are written under the base path. The base path can be set using key {@link #FS_SINK_AUDIT_OUTPUT_PATH_KEY}. * The default path is, * <pre> * <code>System.getProperty("user.dir") + "/lumos_value_audit/local_audit";</code> * </pre> * <li> It uses <code>auditMetadata</code> to build the audit file name and path.<br> * <b>The layout on {@link FileSystem} - </b> * <pre> * |-- &lt;Database&gt; * |-- &lt;Table&gt; * |-- P=&lt;PHASE&gt;.C=&lt;CLUSTER&gt;.E=&lt;EXTRACT_ID&gt;.S=&lt;SNAPSHOT_ID&gt;.D=&lt;DELTA_ID&gt; * |-- *.avro * </pre> * </ul> */ @Alias(value = "FsAuditSink") public class FsAuditSink implements AuditSink { private static final String FS_SINK_AUDIT_OUTPUT_PATH_KEY = "fs.outputDirPath"; private static final String FS_SINK_AUDIT_OUTPUT_DEFAULT_PATH = System.getProperty("user.dir") + "/lumos_value_audit/local_audit"; private static final String FILE_NAME_DELIMITTER = "_"; private final FileSystem fs; private final OutputStream auditFileOutputStream; private final DataFileWriter<GenericRecord> writer; private final Closer closer = Closer.create(); private final ValueAuditRuntimeMetadata auditMetadata; @Getter private final Path auditDirPath; public FsAuditSink(Config config, ValueAuditRuntimeMetadata auditMetadata) throws IOException { this.auditDirPath = new Path(ConfigUtils.getString(config, FS_SINK_AUDIT_OUTPUT_PATH_KEY, FS_SINK_AUDIT_OUTPUT_DEFAULT_PATH)); this.fs = this.auditDirPath.getFileSystem(new Configuration()); this.auditMetadata = auditMetadata; this.auditFileOutputStream = closer.register(fs.create(getAuditFilePath())); DataFileWriter<GenericRecord> dataFileWriter = this.closer.register(new DataFileWriter<GenericRecord>(new GenericDatumWriter<GenericRecord>())); this.writer = this.closer.register(dataFileWriter.create(this.auditMetadata.getTableMetadata().getTableSchema(), this.auditFileOutputStream)); } /** * Returns the complete path of the audit file. Generate the audit file path with format * * <pre> * |-- &lt;Database&gt; * |-- &lt;Table&gt; * |-- P=&lt;PHASE&gt;.C=&lt;CLUSTER&gt;.E=&lt;EXTRACT_ID&gt;.S=&lt;SNAPSHOT_ID&gt;.D=&lt;DELTA_ID&gt; * |-- *.avro * </pre> * */ public Path getAuditFilePath() { StringBuilder auditFileNameBuilder = new StringBuilder(); auditFileNameBuilder.append("P=").append(auditMetadata.getPhase()).append(FILE_NAME_DELIMITTER).append("C=") .append(auditMetadata.getCluster()).append(FILE_NAME_DELIMITTER).append("E=") .append(auditMetadata.getExtractId()).append(FILE_NAME_DELIMITTER).append("S=") .append(auditMetadata.getSnapshotId()).append(FILE_NAME_DELIMITTER).append("D=") .append(auditMetadata.getDeltaId()); return new Path(auditDirPath, PathUtils.combinePaths(auditMetadata.getTableMetadata().getDatabase(), auditMetadata .getTableMetadata().getTable(), auditFileNameBuilder.toString(), auditMetadata.getPartFileName())); } /** * Append this record to the {@link DataFileWriter} * * {@inheritDoc} * @see org.apache.gobblin.audit.values.sink.AuditSink#write(org.apache.avro.generic.GenericRecord) */ @Override public void write(GenericRecord record) throws IOException { this.writer.append(record); } @Override public final void close() throws IOException { this.closer.close(); } }
3,990
0
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/sink/AuditSink.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values.sink; import java.io.Closeable; import java.io.IOException; import org.apache.avro.generic.GenericRecord; /** * An interface for persisting value audits */ public interface AuditSink extends Closeable { /** * Write the <code>record</code> to sink * * @param record to be written * @throws IOException if writing this record failed */ public void write(GenericRecord record) throws IOException; }
3,991
0
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/sink/DefaultAuditSinkFactory.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values.sink; import java.lang.reflect.InvocationTargetException; import lombok.extern.slf4j.Slf4j; import org.apache.commons.lang3.reflect.ConstructorUtils; import com.typesafe.config.Config; import org.apache.gobblin.audit.values.auditor.ValueAuditRuntimeMetadata; import org.apache.gobblin.util.ClassAliasResolver; /** * Default factory class to create new {@link AuditSink}s */ @Slf4j public class DefaultAuditSinkFactory { private static final String AUDIT_SINK_CLASS_NAME_KEY = "class"; private static final String DEFAULT_AUDIT_SINK_CLASS = FsAuditSink.class.getCanonicalName(); private final ClassAliasResolver<AuditSink> aliasResolver; private DefaultAuditSinkFactory() { this.aliasResolver = new ClassAliasResolver<>(AuditSink.class); } /** * Create a new {@link AuditSink} using the alias or cannonical classname specified at {@value #AUDIT_SINK_CLASS_NAME_KEY} in the <code>config</code> * The {@link AuditSink} class MUST have an accessible constructor <code>abc(Config config, TableMetadata tableMetadata)</code> * <br> * If {@value #AUDIT_SINK_CLASS_NAME_KEY} is not set in <code>config</code>, a default {@link #DEFAULT_AUDIT_SINK_CLASS} is used * * @param config job configs * @param auditRuntimeMetadata runtime table metadata * * @return a new instance of {@link AuditSink} */ public AuditSink create(Config config, ValueAuditRuntimeMetadata auditRuntimeMetadata) { String sinkClassName = DEFAULT_AUDIT_SINK_CLASS; if (config.hasPath(AUDIT_SINK_CLASS_NAME_KEY)) { sinkClassName = config.getString(AUDIT_SINK_CLASS_NAME_KEY); } log.info("Using audit sink class name/alias " + sinkClassName); try { return (AuditSink)ConstructorUtils.invokeConstructor(Class.forName(this.aliasResolver.resolve( sinkClassName)), config, auditRuntimeMetadata); } catch (NoSuchMethodException | IllegalAccessException | InvocationTargetException | InstantiationException | ClassNotFoundException e) { throw new RuntimeException(e); } } private static class DefaultAuditSinkFactoryHolder { private static final DefaultAuditSinkFactory INSTANCE = new DefaultAuditSinkFactory(); } public static DefaultAuditSinkFactory getInstance() { return DefaultAuditSinkFactoryHolder.INSTANCE; } }
3,992
0
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/auditor/ValueAuditGenerator.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values.auditor; import java.io.Closeable; import java.io.IOException; import lombok.AllArgsConstructor; import lombok.Getter; import org.apache.avro.generic.GenericRecord; import com.typesafe.config.Config; import org.apache.gobblin.audit.values.policy.column.ColumnProjectionPolicy; import org.apache.gobblin.audit.values.policy.column.DefaultColumnProjectionPolicyFactory; import org.apache.gobblin.audit.values.policy.row.DefaultRowSelectionPolicyFactory; import org.apache.gobblin.audit.values.policy.row.RowSelectionPolicy; import org.apache.gobblin.audit.values.sink.AuditSink; import org.apache.gobblin.audit.values.sink.DefaultAuditSinkFactory; /** * The class that implements value based auditing. The class captures the values of certain * columns from the rows in the dataset using the {@link ColumnProjectionPolicy}. * This is done for every row or for a sample of the rows as defined by the {@link RowSelectionPolicy}. * The selected rows are then written to the {@link AuditSink} * * {@link ValueAuditGenerator#audit(GenericRecord)} is the method that audits an inputRecord. */ @AllArgsConstructor @Getter public class ValueAuditGenerator implements Closeable { public static final String COLUMN_PROJECTION_CONFIG_SCOPE = "columnProjection"; public static final String ROW_SELECTION_CONFIG_SCOPE = "rowSelection"; public static final String AUDIT_SINK_CONFIG_SCOPE = "auditSink"; private final ColumnProjectionPolicy columnProjectionPolicy; private final RowSelectionPolicy rowSelectionPolicy; private final AuditSink auditSink; /** * Factory method to create a new {@link ValueAuditGenerator} * @param config job configs * @param runtimeAuditMetadata is used to pass the table specific runtime information like tablename, databaseName, snapshotName etc. * See {@link ValueAuditRuntimeMetadata} * @return a new {@link ValueAuditGenerator} */ public static ValueAuditGenerator create(Config config, ValueAuditRuntimeMetadata runtimeAuditMetadata) { ColumnProjectionPolicy columnProjectionPolicy = DefaultColumnProjectionPolicyFactory.getInstance().create( config.getConfig(COLUMN_PROJECTION_CONFIG_SCOPE),runtimeAuditMetadata.getTableMetadata()); RowSelectionPolicy rowSelectionPolicy = DefaultRowSelectionPolicyFactory.getInstance().create( config.getConfig(ROW_SELECTION_CONFIG_SCOPE), runtimeAuditMetadata.getTableMetadata(), columnProjectionPolicy); AuditSink auditSink = DefaultAuditSinkFactory.getInstance().create( config.getConfig(AUDIT_SINK_CONFIG_SCOPE), runtimeAuditMetadata); return new ValueAuditGenerator(columnProjectionPolicy, rowSelectionPolicy, auditSink); } /** * Write an audit record for the <code>inputRecord</code> to the {@link AuditSink}. * An audit record is generated for every <code>inputRecord</code> that satisfies the {@link RowSelectionPolicy}. * An audit record is created by projecting <code>inputRecord</code> using the {@link ColumnProjectionPolicy} * * @param inputRecord to be audited * @throws IOException if auditing failed for this record */ public void audit(GenericRecord inputRecord) throws IOException { if (this.rowSelectionPolicy.shouldSelectRow(inputRecord)) { auditSink.write(columnProjectionPolicy.project(inputRecord)); } } @Override public void close() throws IOException { this.auditSink.close(); } }
3,993
0
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/auditor/ValueAuditRuntimeMetadata.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values.auditor; import java.util.List; import lombok.AccessLevel; import lombok.AllArgsConstructor; import lombok.EqualsAndHashCode; import lombok.Getter; import lombok.NonNull; import lombok.RequiredArgsConstructor; import lombok.ToString; import org.apache.avro.Schema; /** * A container for table specific runtime Metadata required for auditing a table. * Use {@link ValueAuditRuntimeMetadataBuilder} to instantiate a new {@link ValueAuditRuntimeMetadata}. * <code>database, table and tableSchema</code> are required fields in the {@link ValueAuditRuntimeMetadata}. * All other fields <code>phase, cluster, extractId, snapshotId, deltaId, partFileName</code> are marked as {@value #DEFAULT_VALUE} * if not present. */ @AllArgsConstructor(access = AccessLevel.PRIVATE) @Getter @ToString @EqualsAndHashCode public class ValueAuditRuntimeMetadata { private static final String DEFAULT_VALUE = "NA"; /** * <i>Required - </i>Table specific metadata like <code>database, table and tableSchema</code> */ private TableMetadata tableMetadata; /** * <i>Optional - </i>The snapshot generation phase being audited */ private Phase phase; /** * <i>Optional - </i>Audited snapshot data's cluster */ private String cluster; /** * <i>Optional - </i>Extract Id of the snapshot */ private String extractId; /** * <i>Optional - </i>Snapshot Id being audited */ private String snapshotId; /** * <i>Optional - </i>Delta Id of the snapshot */ private String deltaId; /** * <i>Optional - </i>Part file in the snapshot being audited */ private String partFileName; public static ValueAuditRuntimeMetadataBuilder builder(String databaseName, String tableName, Schema tableSchema) { return new ValueAuditRuntimeMetadataBuilder(databaseName, tableName, tableSchema); } /** * Container for table specific metadata */ @Getter @RequiredArgsConstructor(access = AccessLevel.PRIVATE) public static class TableMetadata { /** * <i>Required - </i> database name */ @NonNull private String database; /** * <i>Required - </i> table name */ @NonNull private String table; /** * <i>Required - </i> table schema */ @NonNull private Schema tableSchema; /** * <i>Optional - </i> list of key fields in the table that uniquely identify a row. * Each entry in the list is an ordered string specifying the location of the nested key field * For example, field1.nestedField1 refers to the field "nestedField1" inside of field "field1" of the record. */ private List<String> keyFieldLocations; /** * <i>Optional - </i> list of delta fields in the table that are used to track changes in the row over time. * Each entry in the list is an ordered string specifying the location of the nested delta field * For example, field1.nestedField1 refers to the field "nestedField1" inside of field "field1" of the record. */ private List<String> deltaFieldLocations; } /** * An enum for all phases snapshot generation */ public static enum Phase { PULL("Pull from extract"), AVRO_CONV("Convert to avro"), SS_GEN("Snapshot Generation, LSB"), SS_UPD("Snapshot update, VSB"), SS_MAT("Snapshot materialization, QSB"), SS_PUB("Publish Snapshot"), NA("Not Applicable"); private String description; Phase(String description) { this.description = description; } public String getDescription() { return this.description; } } /** * Builder to build A {@link ValueAuditRuntimeMetadata}, <code>databaseName, tableName and tableSchema</code> are required */ public static class ValueAuditRuntimeMetadataBuilder { private TableMetadata tableMetadata; private Phase phase = Phase.NA; private String cluster = DEFAULT_VALUE; private String extractId = DEFAULT_VALUE; private String snapshotId = DEFAULT_VALUE; private String deltaId = DEFAULT_VALUE; private String partFileName = DEFAULT_VALUE; public ValueAuditRuntimeMetadataBuilder(String databaseName, String tableName, Schema tableSchema) { this.tableMetadata = new TableMetadata(databaseName, tableName, tableSchema); } public ValueAuditRuntimeMetadataBuilder phase(final Phase phase) { this.phase = phase; return this; } public ValueAuditRuntimeMetadataBuilder cluster(final String cluster) { this.cluster = cluster; return this; } public ValueAuditRuntimeMetadataBuilder extractId(final String extractId) { this.extractId = extractId; return this; } public ValueAuditRuntimeMetadataBuilder snapshotId(final String snapshotId) { this.snapshotId = snapshotId; return this; } public ValueAuditRuntimeMetadataBuilder deltaId(final String deltaId) { this.deltaId = deltaId; return this; } public ValueAuditRuntimeMetadataBuilder partFileName(final String partFileName) { this.partFileName = partFileName; return this; } public ValueAuditRuntimeMetadataBuilder tableMetadataKeyFieldLocations(List<String> keyFieldLocations) { this.tableMetadata.keyFieldLocations = keyFieldLocations; return this; } public ValueAuditRuntimeMetadataBuilder tableMetadataDeltaFieldLocations(List<String> deltaFieldLocations) { this.tableMetadata.keyFieldLocations = deltaFieldLocations; return this; } public ValueAuditRuntimeMetadata build() { return new ValueAuditRuntimeMetadata(tableMetadata, phase, cluster, extractId, snapshotId, deltaId, partFileName); } } }
3,994
0
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/policy
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/policy/column/ColumnProjectionPolicy.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values.policy.column; import java.util.List; import org.apache.avro.generic.GenericRecord; /** * An interface that projects certain columns/fields of an input {@link GenericRecord} to generate a new {@link GenericRecord} that can be audited. * The field locations to project are an ordered string specifying the location of the nested field to retrieve. * For example, field1.nestedField1 takes the the value of the field "field1" of the record, and retrieves the field "nestedField1" from it */ public interface ColumnProjectionPolicy { /** * Get the key aka unique identifier fields to project in a {@link GenericRecord} */ public List<String> getKeyColumnsToProject(); /** * Get the delta fields to project in a {@link GenericRecord}. These are the fields used to determine changes over time. */ public List<String> getDeltaColumnsToProject(); /** * A union of {@link #getKeyColumnsToProject()} and {@link #getDeltaColumnsToProject()} */ public List<String> getAllColumnsToProject(); /** * Project key and delta columns/fields of the <code>inputRecord</code> and return a new {@link GenericRecord} with only the projected columns/fields * @param inputRecord the original record with all columns/fields * @return a new {@link GenericRecord} with only {@link #getAllColumnsToProject()} */ public GenericRecord project(GenericRecord inputRecord); }
3,995
0
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/policy
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/policy/column/DefaultColumnProjectionPolicyFactory.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values.policy.column; import java.lang.reflect.InvocationTargetException; import lombok.extern.slf4j.Slf4j; import org.apache.commons.lang3.reflect.ConstructorUtils; import com.google.common.base.Preconditions; import com.typesafe.config.Config; import org.apache.gobblin.audit.values.auditor.ValueAuditRuntimeMetadata; import org.apache.gobblin.util.ClassAliasResolver; /** * Default factory class to create new {@link ColumnProjectionPolicy}s */ @Slf4j public class DefaultColumnProjectionPolicyFactory { private static final String COLUMN_PROJECTION_POLICY_CLASS_NAME_KEY = "class"; private final ClassAliasResolver<ColumnProjectionPolicy> aliasResolver; private DefaultColumnProjectionPolicyFactory() { this.aliasResolver = new ClassAliasResolver<>(ColumnProjectionPolicy.class); } /** * Create a new {@link ColumnProjectionPolicy} using the alias or cannonical classname specified at {@value #COLUMN_PROJECTION_POLICY_CLASS_NAME_KEY} in the <code>config</code> * The {@link ColumnProjectionPolicy} class MUST have an accessible constructor <code>abc(Config config, TableMetadata tableMetadata)</code> * <b>Note : Must have the key {@value #COLUMN_PROJECTION_POLICY_CLASS_NAME_KEY} set in <code>config</code> to create the {@link ColumnProjectionPolicy}</b> * * @param config job configs, Must have the key {@value #COLUMN_PROJECTION_POLICY_CLASS_NAME_KEY} set to create the {@link ColumnProjectionPolicy} * @param tableMetadata runtime table metadata * * @return a new instance of {@link ColumnProjectionPolicy} */ public ColumnProjectionPolicy create(Config config, ValueAuditRuntimeMetadata.TableMetadata tableMetadata) { Preconditions.checkArgument(config.hasPath(COLUMN_PROJECTION_POLICY_CLASS_NAME_KEY)); log.info("Using column projection class name/alias " + config.getString(COLUMN_PROJECTION_POLICY_CLASS_NAME_KEY)); try { return (ColumnProjectionPolicy)ConstructorUtils.invokeConstructor(Class.forName(this.aliasResolver.resolve( config.getString(COLUMN_PROJECTION_POLICY_CLASS_NAME_KEY))), config, tableMetadata); } catch (NoSuchMethodException | IllegalAccessException | InvocationTargetException | InstantiationException | ClassNotFoundException e) { throw new RuntimeException(e); } } private static class DefaultColumnProjectionPolicyFactoryHolder { private static final DefaultColumnProjectionPolicyFactory INSTANCE = new DefaultColumnProjectionPolicyFactory(); } public static DefaultColumnProjectionPolicyFactory getInstance() { return DefaultColumnProjectionPolicyFactoryHolder.INSTANCE; } }
3,996
0
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/policy
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/policy/column/ProjectAllColumnProjectionPolicy.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values.policy.column; import java.util.List; import org.apache.avro.generic.GenericRecord; import com.typesafe.config.Config; import org.apache.gobblin.annotation.Alias; import org.apache.gobblin.audit.values.auditor.ValueAuditRuntimeMetadata; /** * An {@link AbstractColumnProjectionPolicy} that projects all columns/fields of the <code>inputRecord</code> */ @Alias(value = "ProjectAll") public class ProjectAllColumnProjectionPolicy extends AbstractColumnProjectionPolicy { public ProjectAllColumnProjectionPolicy(Config config, ValueAuditRuntimeMetadata.TableMetadata tableMetadata) { super(config, tableMetadata); } /** * Return the entire <code>inputRecord</code>. All fields are projected * * {@inheritDoc} * @see org.apache.gobblin.audit.values.policy.column.ColumnProjectionPolicy#project(org.apache.avro.generic.GenericRecord) */ @Override public GenericRecord project(GenericRecord inputRecord) { return inputRecord; } @Override public List<String> getKeyColumnsToProject() { return this.tableMetadata.getKeyFieldLocations(); } @Override public List<String> getDeltaColumnsToProject() { return this.tableMetadata.getDeltaFieldLocations(); } }
3,997
0
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/policy
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/policy/column/AbstractColumnProjectionPolicy.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values.policy.column; import java.util.List; import org.apache.avro.Schema; import com.google.common.collect.ImmutableList; import com.typesafe.config.Config; import org.apache.gobblin.audit.values.auditor.ValueAuditRuntimeMetadata; /** * A base {@link ColumnProjectionPolicy} that reads <code>config</code> to initialize the key and delta columns to project for a table. * <ul> * <li> Key and delta fields/column locations to project should be provided by concrete subclasses by implementing {@link #getDeltaColumnsToProject()} * and {@link #getKeyColumnsToProject()} * <li> The protected member <code>tableMetadata</code> contains the {@link Schema}, tableName and databaseName to derive the projection columns * </ul> */ public abstract class AbstractColumnProjectionPolicy implements ColumnProjectionPolicy { protected final ValueAuditRuntimeMetadata.TableMetadata tableMetadata; public AbstractColumnProjectionPolicy(Config config, ValueAuditRuntimeMetadata.TableMetadata tableMetadata) { this.tableMetadata = tableMetadata; } /** * Combine both key columns and delta columns to project * {@inheritDoc} * @see org.apache.gobblin.audit.values.policy.column.ColumnProjectionPolicy#getAllColumnsToProject() */ public List<String> getAllColumnsToProject() { return ImmutableList.<String> builder().addAll(getKeyColumnsToProject()).addAll(getDeltaColumnsToProject()).build(); } }
3,998
0
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/policy
Create_ds/gobblin/gobblin-audit/src/main/java/org/apache/gobblin/audit/values/policy/row/RowSelectionPolicy.java
/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.apache.gobblin.audit.values.policy.row; import org.apache.avro.generic.GenericRecord; /** * An interface to decide if a row needs to be audited */ public interface RowSelectionPolicy { /** * Finds if this <code>genericRecord</code> needs to be audited * * @param inputRecord to be audited * @return <code>true</code> if <code>inputRecord</code> needs to be audited <code>false</code> otherwise */ public boolean shouldSelectRow(GenericRecord inputRecord); }
3,999