Spaces:
Running
Running
File size: 11,054 Bytes
a686b1b |
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 |
"""
Sistema de observabilidade com Prometheus e OpenTelemetry.
Metricas:
- Contadores de requests
- Histogramas de latencia
- Gauges de estado do sistema
- Metricas customizadas do RAG
"""
from typing import Dict, Any, Optional
import time
from functools import wraps
from contextlib import contextmanager
try:
from prometheus_client import (
Counter,
Histogram,
Gauge,
generate_latest,
CONTENT_TYPE_LATEST,
CollectorRegistry
)
PROMETHEUS_AVAILABLE = True
except ImportError:
PROMETHEUS_AVAILABLE = False
print("Aviso: prometheus_client nao instalado. Metricas desabilitadas.")
try:
from opentelemetry import trace
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.resources import Resource
from opentelemetry.exporter.otlp.proto.grpc.trace_exporter import OTLPSpanExporter
from opentelemetry.sdk.trace.export import BatchSpanProcessor
OTEL_AVAILABLE = True
except ImportError:
OTEL_AVAILABLE = False
print("Aviso: opentelemetry nao instalado. Traces desabilitados.")
class MetricsCollector:
"""Coletor de metricas Prometheus."""
def __init__(self, registry: Optional['CollectorRegistry'] = None):
"""
Inicializa coletor.
Args:
registry: Registry do Prometheus (opcional)
"""
if not PROMETHEUS_AVAILABLE:
self.enabled = False
return
self.enabled = True
self.registry = registry
# Contadores
self.requests_total = Counter(
'rag_requests_total',
'Total de requests ao sistema',
['endpoint', 'method'],
registry=registry
)
self.queries_total = Counter(
'rag_queries_total',
'Total de queries RAG',
registry=registry
)
self.documents_ingested_total = Counter(
'rag_documents_ingested_total',
'Total de documentos ingeridos',
registry=registry
)
self.chunks_created_total = Counter(
'rag_chunks_created_total',
'Total de chunks criados',
registry=registry
)
self.errors_total = Counter(
'rag_errors_total',
'Total de erros',
['error_type'],
registry=registry
)
# Histogramas de latencia
self.query_latency = Histogram(
'rag_query_latency_seconds',
'Latencia de queries RAG',
buckets=[0.1, 0.5, 1.0, 2.0, 5.0, 10.0],
registry=registry
)
self.embedding_latency = Histogram(
'rag_embedding_latency_seconds',
'Latencia de geracao de embeddings',
buckets=[0.05, 0.1, 0.25, 0.5, 1.0],
registry=registry
)
self.generation_latency = Histogram(
'rag_generation_latency_seconds',
'Latencia de geracao de respostas',
buckets=[0.5, 1.0, 2.0, 5.0, 10.0, 20.0],
registry=registry
)
self.db_query_latency = Histogram(
'rag_db_query_latency_seconds',
'Latencia de queries ao banco',
buckets=[0.01, 0.05, 0.1, 0.25, 0.5],
registry=registry
)
# Gauges
self.active_connections = Gauge(
'rag_active_connections',
'Conexoes ativas ao banco',
registry=registry
)
self.cache_size = Gauge(
'rag_cache_size_bytes',
'Tamanho do cache de embeddings',
registry=registry
)
self.documents_count = Gauge(
'rag_documents_count',
'Numero total de documentos',
registry=registry
)
self.chunks_count = Gauge(
'rag_chunks_count',
'Numero total de chunks',
registry=registry
)
def increment_requests(self, endpoint: str, method: str):
"""Incrementa contador de requests."""
if self.enabled:
self.requests_total.labels(endpoint=endpoint, method=method).inc()
def increment_queries(self):
"""Incrementa contador de queries."""
if self.enabled:
self.queries_total.inc()
def increment_documents_ingested(self, count: int = 1):
"""Incrementa contador de documentos ingeridos."""
if self.enabled:
self.documents_ingested_total.inc(count)
def increment_chunks_created(self, count: int):
"""Incrementa contador de chunks criados."""
if self.enabled:
self.chunks_created_total.inc(count)
def increment_errors(self, error_type: str):
"""Incrementa contador de erros."""
if self.enabled:
self.errors_total.labels(error_type=error_type).inc()
@contextmanager
def measure_query_latency(self):
"""Context manager para medir latencia de query."""
start = time.time()
try:
yield
finally:
if self.enabled:
self.query_latency.observe(time.time() - start)
@contextmanager
def measure_embedding_latency(self):
"""Context manager para medir latencia de embedding."""
start = time.time()
try:
yield
finally:
if self.enabled:
self.embedding_latency.observe(time.time() - start)
@contextmanager
def measure_generation_latency(self):
"""Context manager para medir latencia de geracao."""
start = time.time()
try:
yield
finally:
if self.enabled:
self.generation_latency.observe(time.time() - start)
@contextmanager
def measure_db_query_latency(self):
"""Context manager para medir latencia de query ao banco."""
start = time.time()
try:
yield
finally:
if self.enabled:
self.db_query_latency.observe(time.time() - start)
def set_active_connections(self, count: int):
"""Define numero de conexoes ativas."""
if self.enabled:
self.active_connections.set(count)
def set_cache_size(self, size_bytes: int):
"""Define tamanho do cache."""
if self.enabled:
self.cache_size.set(size_bytes)
def set_documents_count(self, count: int):
"""Define numero de documentos."""
if self.enabled:
self.documents_count.set(count)
def set_chunks_count(self, count: int):
"""Define numero de chunks."""
if self.enabled:
self.chunks_count.set(count)
def get_metrics(self) -> str:
"""
Retorna metricas em formato Prometheus.
Returns:
Metricas formatadas
"""
if not self.enabled:
return ""
return generate_latest(self.registry).decode('utf-8')
class TracingManager:
"""Gerenciador de traces OpenTelemetry."""
def __init__(
self,
service_name: str = "rag-template",
otlp_endpoint: Optional[str] = None
):
"""
Inicializa tracing.
Args:
service_name: Nome do servico
otlp_endpoint: Endpoint OTLP (opcional)
"""
if not OTEL_AVAILABLE:
self.enabled = False
return
self.enabled = True
self.service_name = service_name
# Configurar resource
resource = Resource.create({"service.name": service_name})
# Configurar tracer provider
provider = TracerProvider(resource=resource)
# Adicionar exporter se endpoint fornecido
if otlp_endpoint:
exporter = OTLPSpanExporter(endpoint=otlp_endpoint)
processor = BatchSpanProcessor(exporter)
provider.add_span_processor(processor)
# Configurar trace provider global
trace.set_tracer_provider(provider)
# Obter tracer
self.tracer = trace.get_tracer(__name__)
@contextmanager
def trace_operation(self, operation_name: str, attributes: Optional[Dict[str, Any]] = None):
"""
Context manager para criar span.
Args:
operation_name: Nome da operacao
attributes: Atributos do span (opcional)
"""
if not self.enabled:
yield
return
with self.tracer.start_as_current_span(operation_name) as span:
if attributes:
for key, value in attributes.items():
span.set_attribute(key, value)
yield span
# Instancia global
_metrics_collector = None
_tracing_manager = None
def get_metrics_collector(registry: Optional['CollectorRegistry'] = None) -> MetricsCollector:
"""
Obtem instancia global do coletor de metricas.
Args:
registry: Registry do Prometheus (opcional)
Returns:
Coletor de metricas
"""
global _metrics_collector
if _metrics_collector is None:
_metrics_collector = MetricsCollector(registry=registry)
return _metrics_collector
def get_tracing_manager(
service_name: str = "rag-template",
otlp_endpoint: Optional[str] = None
) -> TracingManager:
"""
Obtem instancia global do gerenciador de tracing.
Args:
service_name: Nome do servico
otlp_endpoint: Endpoint OTLP (opcional)
Returns:
Gerenciador de tracing
"""
global _tracing_manager
if _tracing_manager is None:
_tracing_manager = TracingManager(
service_name=service_name,
otlp_endpoint=otlp_endpoint
)
return _tracing_manager
# Decoradores
def track_query_latency(func):
"""Decorator para rastrear latencia de queries."""
@wraps(func)
def wrapper(*args, **kwargs):
metrics = get_metrics_collector()
with metrics.measure_query_latency():
result = func(*args, **kwargs)
metrics.increment_queries()
return result
return wrapper
def track_embedding_latency(func):
"""Decorator para rastrear latencia de embeddings."""
@wraps(func)
def wrapper(*args, **kwargs):
metrics = get_metrics_collector()
with metrics.measure_embedding_latency():
return func(*args, **kwargs)
return wrapper
def track_generation_latency(func):
"""Decorator para rastrear latencia de geracao."""
@wraps(func)
def wrapper(*args, **kwargs):
metrics = get_metrics_collector()
with metrics.measure_generation_latency():
return func(*args, **kwargs)
return wrapper
def trace_operation(operation_name: str):
"""Decorator para criar span OpenTelemetry."""
def decorator(func):
@wraps(func)
def wrapper(*args, **kwargs):
tracing = get_tracing_manager()
with tracing.trace_operation(operation_name):
return func(*args, **kwargs)
return wrapper
return decorator
|