File size: 15,827 Bytes
90c6b42 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 | """
ATOM Communication Memory Production Monitoring System
Real-time monitoring, alerting, and performance tracking
"""
import asyncio
from dataclasses import asdict, dataclass
from datetime import datetime, timedelta
import json
import logging
import time
from typing import Any, Dict, List, Optional
from integrations.atom_communication_ingestion_pipeline import ingestion_pipeline, memory_manager
logger = logging.getLogger(__name__)
@dataclass
class MonitoringMetric:
"""Monitoring metric data structure"""
name: str
value: float
unit: str
timestamp: datetime
tags: Dict[str, str]
threshold: Optional[float] = None
@dataclass
class Alert:
"""Alert data structure"""
id: str
severity: str # info, warning, error, critical
title: str
message: str
timestamp: datetime
resolved: bool = False
resolved_at: Optional[datetime] = None
tags: Dict[str, str]
class AtomCommunicationMemoryMonitoring:
"""Production monitoring system for ATOM communication memory"""
def __init__(self):
self.metrics: List[MonitoringMetric] = []
self.alerts: List[Alert] = []
self.is_running = False
self.monitoring_interval = 60 # seconds
self.alert_thresholds = {
'ingestion_rate': 0.1, # messages per second
'error_rate': 0.05, # 5% error rate
'memory_usage': 0.8, # 80% memory usage
'search_latency': 1.0, # 1 second
'database_size': 100_000_000_000 # 100GB
}
async def start_monitoring(self):
"""Start the monitoring system"""
self.is_running = True
logger.info("Starting ATOM communication memory monitoring")
while self.is_running:
try:
await self.collect_metrics()
await self.check_alerts()
await asyncio.sleep(self.monitoring_interval)
except Exception as e:
logger.error(f"Error in monitoring loop: {str(e)}")
await asyncio.sleep(60) # Wait longer on error
def stop_monitoring(self):
"""Stop the monitoring system"""
self.is_running = False
logger.info("Stopping ATOM communication memory monitoring")
async def collect_metrics(self):
"""Collect monitoring metrics"""
try:
timestamp = datetime.now()
# Get ingestion stats
stats = ingestion_pipeline.get_ingestion_stats()
# Database metrics
db_metrics = await self._collect_database_metrics(timestamp)
# Ingestion metrics
ingestion_metrics = await self._collect_ingestion_metrics(stats, timestamp)
# Performance metrics
performance_metrics = await self._collect_performance_metrics(timestamp)
# Add all metrics
self.metrics.extend(db_metrics + ingestion_metrics + performance_metrics)
# Keep only last 24 hours of metrics
cutoff_time = timestamp - timedelta(hours=24)
self.metrics = [m for m in self.metrics if m.timestamp > cutoff_time]
logger.info(f"Collected {len(db_metrics + ingestion_metrics + performance_metrics)} metrics")
except Exception as e:
logger.error(f"Error collecting metrics: {str(e)}")
async def _collect_database_metrics(self, timestamp: datetime) -> List[MonitoringMetric]:
"""Collect database-related metrics"""
metrics = []
try:
if memory_manager.connections_table:
# Get record count
df = memory_manager.connections_table.to_pandas()
record_count = len(df)
metrics.append(MonitoringMetric(
name="database_record_count",
value=record_count,
unit="records",
timestamp=timestamp,
tags={"table": "atom_communications"},
threshold=self.alert_thresholds['database_size']
))
# Get database size (estimated)
estimated_size = record_count * 1024 # Estimate 1KB per record
metrics.append(MonitoringMetric(
name="database_size",
value=estimated_size,
unit="bytes",
timestamp=timestamp,
tags={"table": "atom_communications"},
threshold=self.alert_thresholds['database_size']
))
# App distribution
app_dist = df["app_type"].value_counts().to_dict()
for app, count in app_dist.items():
metrics.append(MonitoringMetric(
name=f"records_{app}",
value=count,
unit="records",
timestamp=timestamp,
tags={"app": app, "metric": "record_count"}
))
except Exception as e:
logger.error(f"Error collecting database metrics: {str(e)}")
return metrics
async def _collect_ingestion_metrics(self, stats: Dict[str, Any], timestamp: datetime) -> List[MonitoringMetric]:
"""Collect ingestion-related metrics"""
metrics = []
try:
# Total messages
total_messages = stats.get('total_messages', 0)
metrics.append(MonitoringMetric(
name="total_messages_ingested",
value=total_messages,
unit="messages",
timestamp=timestamp,
tags={"metric": "total_ingestion"}
))
# Active streams
active_streams = len(stats.get('active_streams', []))
metrics.append(MonitoringMetric(
name="active_real_time_streams",
value=active_streams,
unit="streams",
timestamp=timestamp,
tags={"metric": "active_streams"}
))
# Configured apps
configured_apps = len(stats.get('configured_apps', []))
metrics.append(MonitoringMetric(
name="configured_apps",
value=configured_apps,
unit="apps",
timestamp=timestamp,
tags={"metric": "configured_apps"}
))
except Exception as e:
logger.error(f"Error collecting ingestion metrics: {str(e)}")
return metrics
async def _collect_performance_metrics(self, timestamp: datetime) -> List[MonitoringMetric]:
"""Collect performance-related metrics"""
metrics = []
try:
# Ingestion rate (simplified)
recent_metrics = [m for m in self.metrics
if m.name == "total_messages_ingested"
and (timestamp - m.timestamp).total_seconds() < 300] # Last 5 minutes
if len(recent_metrics) >= 2:
recent_metrics.sort(key=lambda x: x.timestamp)
latest_count = recent_metrics[-1].value
earliest_count = recent_metrics[0].value
time_diff = (recent_metrics[-1].timestamp - recent_metrics[0].timestamp).total_seconds()
if time_diff > 0:
ingestion_rate = (latest_count - earliest_count) / time_diff
metrics.append(MonitoringMetric(
name="ingestion_rate",
value=ingestion_rate,
unit="messages/second",
timestamp=timestamp,
tags={"metric": "performance"},
threshold=self.alert_thresholds['ingestion_rate']
))
# Memory usage (simplified - would need actual monitoring)
import psutil
memory_percent = psutil.virtual_memory().percent / 100
metrics.append(MonitoringMetric(
name="memory_usage",
value=memory_percent,
unit="fraction",
timestamp=timestamp,
tags={"metric": "performance"},
threshold=self.alert_thresholds['memory_usage']
))
except Exception as e:
logger.error(f"Error collecting performance metrics: {str(e)}")
return metrics
async def check_alerts(self):
"""Check thresholds and generate alerts"""
try:
timestamp = datetime.now()
# Get latest metrics for each metric name
latest_metrics = {}
for metric in self.metrics:
if metric.name not in latest_metrics or metric.timestamp > latest_metrics[metric.name].timestamp:
latest_metrics[metric.name] = metric
# Check thresholds
for metric_name, metric in latest_metrics.items():
if metric.threshold and metric.value > metric.threshold:
await self._create_alert(
severity="warning",
title=f"Threshold exceeded for {metric_name}",
message=f"{metric_name}: {metric.value:.2f} {metric.unit} (threshold: {metric.threshold})",
timestamp=timestamp,
tags=metric.tags
)
# Check for system health
if not memory_manager.db:
await self._create_alert(
severity="critical",
title="Database connection lost",
message="LanceDB database connection is not available",
timestamp=timestamp,
tags={"component": "database"}
)
except Exception as e:
logger.error(f"Error checking alerts: {str(e)}")
async def _create_alert(self, severity: str, title: str, message: str,
timestamp: datetime, tags: Dict[str, str]):
"""Create a new alert"""
alert_id = f"alert_{int(timestamp.timestamp())}_{len(self.alerts)}"
# Check if similar alert already exists
existing_alert = next((a for a in self.alerts if not a.resolved and a.title == title), None)
if existing_alert:
# Update existing alert
existing_alert.timestamp = timestamp
existing_alert.message = message
else:
# Create new alert
alert = Alert(
id=alert_id,
severity=severity,
title=title,
message=message,
timestamp=timestamp,
tags=tags
)
self.alerts.append(alert)
logger.warning(f"Alert created: {severity} - {title}")
def get_metrics_summary(self, time_window: int = 3600) -> Dict[str, Any]:
"""Get summary of metrics for the last N seconds"""
try:
cutoff_time = datetime.now() - timedelta(seconds=time_window)
recent_metrics = [m for m in self.metrics if m.timestamp > cutoff_time]
# Group metrics by name
metrics_by_name = {}
for metric in recent_metrics:
if metric.name not in metrics_by_name:
metrics_by_name[metric.name] = []
metrics_by_name[metric.name].append(metric)
# Calculate summaries
summary = {
"time_window": time_window,
"metric_count": len(recent_metrics),
"metrics": {}
}
for name, metric_list in metrics_by_name.items():
values = [m.value for m in metric_list]
summary["metrics"][name] = {
"latest": values[-1] if values else None,
"average": sum(values) / len(values) if values else None,
"min": min(values) if values else None,
"max": max(values) if values else None,
"count": len(values),
"unit": metric_list[0].unit if metric_list else None
}
return summary
except Exception as e:
logger.error(f"Error getting metrics summary: {str(e)}")
return {"error": str(e)}
def get_alerts_summary(self, include_resolved: bool = False) -> Dict[str, Any]:
"""Get summary of alerts"""
try:
alerts = self.alerts if include_resolved else [a for a in self.alerts if not a.resolved]
# Count by severity
severity_counts = {}
for alert in alerts:
severity_counts[alert.severity] = severity_counts.get(alert.severity, 0) + 1
return {
"total_alerts": len(alerts),
"unresolved_alerts": len([a for a in alerts if not a.resolved]),
"severity_distribution": severity_counts,
"recent_alerts": [
{
"id": alert.id,
"severity": alert.severity,
"title": alert.title,
"message": alert.message,
"timestamp": alert.timestamp.isoformat(),
"resolved": alert.resolved
}
for alert in sorted(alerts, key=lambda x: x.timestamp, reverse=True)[:10]
]
}
except Exception as e:
logger.error(f"Error getting alerts summary: {str(e)}")
return {"error": str(e)}
def get_health_status(self) -> Dict[str, Any]:
"""Get overall system health status"""
try:
# Check critical components
health_checks = {
"database": memory_manager.db is not None,
"ingestion_pipeline": len(ingestion_pipeline.ingestion_configs) > 0,
"monitoring": self.is_running
}
# Check recent errors
recent_alerts = [a for a in self.alerts
if not a.resolved
and a.severity in ["error", "critical"]
and (datetime.now() - a.timestamp).total_seconds() < 3600]
overall_status = "healthy"
if not all(health_checks.values()):
overall_status = "unhealthy"
elif recent_alerts:
overall_status = "degraded"
return {
"overall_status": overall_status,
"timestamp": datetime.now().isoformat(),
"health_checks": health_checks,
"recent_critical_alerts": len(recent_alerts),
"monitoring_active": self.is_running
}
except Exception as e:
logger.error(f"Error getting health status: {str(e)}")
return {"error": str(e), "overall_status": "unknown"}
# Create global monitoring instance
atom_memory_monitoring = AtomCommunicationMemoryMonitoring()
# Export for use
__all__ = [
'AtomCommunicationMemoryMonitoring',
'atom_memory_monitoring',
'MonitoringMetric',
'Alert'
]
|