File size: 23,226 Bytes
c97e35f |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 |
"""
SLA/SLO monitoring system for tracking service level objectives.
This module provides comprehensive SLA/SLO tracking with automated
alerting and reporting capabilities.
"""
import asyncio
import time
from typing import Dict, Any, List, Optional, Callable, Union
from datetime import datetime, timedelta
from enum import Enum
from dataclasses import dataclass, field
from collections import deque, defaultdict
import statistics
from src.core import get_logger
from src.infrastructure.observability import BusinessMetrics, get_structured_logger
logger = get_structured_logger(__name__, component="slo_monitor")
class SLOType(str, Enum):
"""Types of SLO metrics."""
AVAILABILITY = "availability"
LATENCY = "latency"
ERROR_RATE = "error_rate"
THROUGHPUT = "throughput"
CUSTOM = "custom"
class AlertSeverity(str, Enum):
"""Alert severity levels."""
INFO = "info"
WARNING = "warning"
CRITICAL = "critical"
class TimeWindow(str, Enum):
"""Time windows for SLO calculations."""
MINUTE_1 = "1m"
MINUTE_5 = "5m"
MINUTE_15 = "15m"
HOUR_1 = "1h"
HOUR_4 = "4h"
HOUR_24 = "24h"
DAY_7 = "7d"
DAY_30 = "30d"
@dataclass
class SLOTarget:
"""SLO target definition."""
name: str
slo_type: SLOType
target_value: float # Target threshold (e.g., 99.9 for 99.9% availability)
time_window: TimeWindow
description: str
warning_threshold: float = 95.0 # % of target that triggers warning
critical_threshold: float = 90.0 # % of target that triggers critical alert
def get_warning_value(self) -> float:
"""Get warning threshold value."""
return self.target_value * (self.warning_threshold / 100)
def get_critical_value(self) -> float:
"""Get critical threshold value."""
return self.target_value * (self.critical_threshold / 100)
@dataclass
class SLOMetric:
"""Single SLO metric measurement."""
timestamp: datetime
value: float
success: bool = True
metadata: Dict[str, Any] = field(default_factory=dict)
@dataclass
class SLOViolation:
"""SLO violation record."""
slo_name: str
violation_time: datetime
actual_value: float
target_value: float
severity: AlertSeverity
duration_minutes: float
details: Dict[str, Any] = field(default_factory=dict)
class SLOCalculator:
"""Calculator for different types of SLO metrics."""
@staticmethod
def calculate_availability(metrics: List[SLOMetric]) -> float:
"""Calculate availability percentage."""
if not metrics:
return 0.0
successful = sum(1 for m in metrics if m.success)
total = len(metrics)
return (successful / total) * 100
@staticmethod
def calculate_latency_percentile(metrics: List[SLOMetric], percentile: float = 95.0) -> float:
"""Calculate latency percentile."""
if not metrics:
return 0.0
values = [m.value for m in metrics]
return statistics.quantiles(values, n=100)[int(percentile) - 1] if len(values) > 1 else values[0]
@staticmethod
def calculate_error_rate(metrics: List[SLOMetric]) -> float:
"""Calculate error rate percentage."""
if not metrics:
return 0.0
errors = sum(1 for m in metrics if not m.success)
total = len(metrics)
return (errors / total) * 100
@staticmethod
def calculate_throughput(metrics: List[SLOMetric], time_window_minutes: float) -> float:
"""Calculate throughput (requests per minute)."""
if not metrics or time_window_minutes <= 0:
return 0.0
return len(metrics) / time_window_minutes
@staticmethod
def calculate_custom(metrics: List[SLOMetric], calculation_func: Callable) -> float:
"""Calculate custom metric using provided function."""
return calculation_func(metrics)
class SLOMonitor:
"""
Monitor for tracking SLA/SLO compliance.
Features:
- Real-time SLO tracking
- Automated violation detection
- Alert generation
- Historical trend analysis
- Error budget tracking
"""
def __init__(self, max_history_size: int = 10000):
"""
Initialize SLO monitor.
Args:
max_history_size: Maximum number of metrics to keep in memory
"""
self.max_history_size = max_history_size
# SLO definitions
self.slo_targets: Dict[str, SLOTarget] = {}
# Metric storage
self.metrics: Dict[str, deque] = defaultdict(lambda: deque(maxlen=max_history_size))
# Violation tracking
self.violations: Dict[str, List[SLOViolation]] = defaultdict(list)
self.current_violations: Dict[str, SLOViolation] = {}
# Alert callbacks
self.alert_callbacks: List[Callable] = []
# Calculator
self.calculator = SLOCalculator()
# Statistics
self.stats = {
"total_measurements": 0,
"total_violations": 0,
"alerts_sent": 0
}
self.logger = get_structured_logger(__name__, component="slo_monitor")
# Initialize default SLOs
self._initialize_default_slos()
def _initialize_default_slos(self):
"""Initialize default SLO targets for Cidadão.AI."""
# API Availability
self.register_slo(SLOTarget(
name="api_availability",
slo_type=SLOType.AVAILABILITY,
target_value=99.9, # 99.9% uptime
time_window=TimeWindow.HOUR_24,
description="API endpoint availability over 24 hours",
warning_threshold=98.0,
critical_threshold=95.0
))
# API Response Time
self.register_slo(SLOTarget(
name="api_latency_p95",
slo_type=SLOType.LATENCY,
target_value=2000.0, # 2 seconds P95
time_window=TimeWindow.HOUR_1,
description="95th percentile API response time under 2 seconds",
warning_threshold=90.0,
critical_threshold=80.0
))
# Investigation Success Rate
self.register_slo(SLOTarget(
name="investigation_success_rate",
slo_type=SLOType.AVAILABILITY,
target_value=95.0, # 95% success rate
time_window=TimeWindow.HOUR_4,
description="Investigation completion success rate",
warning_threshold=92.0,
critical_threshold=88.0
))
# Agent Task Error Rate
self.register_slo(SLOTarget(
name="agent_error_rate",
slo_type=SLOType.ERROR_RATE,
target_value=1.0, # < 1% error rate
time_window=TimeWindow.HOUR_1,
description="Agent task error rate under 1%",
warning_threshold=80.0,
critical_threshold=60.0
))
# Database Query Performance
self.register_slo(SLOTarget(
name="database_latency_p90",
slo_type=SLOType.LATENCY,
target_value=500.0, # 500ms P90
time_window=TimeWindow.MINUTE_15,
description="90th percentile database query time under 500ms",
warning_threshold=85.0,
critical_threshold=70.0
))
# Anomaly Detection Accuracy
self.register_slo(SLOTarget(
name="anomaly_detection_accuracy",
slo_type=SLOType.CUSTOM,
target_value=90.0, # 90% accuracy
time_window=TimeWindow.HOUR_24,
description="Anomaly detection accuracy rate",
warning_threshold=95.0,
critical_threshold=90.0
))
def register_slo(self, slo_target: SLOTarget):
"""Register an SLO target for monitoring."""
self.slo_targets[slo_target.name] = slo_target
self.logger.info(
f"Registered SLO: {slo_target.name}",
operation="register_slo",
slo_type=slo_target.slo_type.value,
target_value=slo_target.target_value,
time_window=slo_target.time_window.value
)
def register_alert_callback(self, callback: Callable[[SLOViolation], None]):
"""Register callback for SLO violation alerts."""
self.alert_callbacks.append(callback)
def record_metric(
self,
slo_name: str,
value: float,
success: bool = True,
metadata: Optional[Dict[str, Any]] = None
):
"""
Record a metric for SLO monitoring.
Args:
slo_name: Name of the SLO
value: Metric value
success: Whether the operation was successful
metadata: Additional metadata
"""
if slo_name not in self.slo_targets:
self.logger.warning(f"Unknown SLO: {slo_name}")
return
metric = SLOMetric(
timestamp=datetime.utcnow(),
value=value,
success=success,
metadata=metadata or {}
)
self.metrics[slo_name].append(metric)
self.stats["total_measurements"] += 1
# Check for violations
asyncio.create_task(self._check_slo_compliance(slo_name))
self.logger.debug(
f"Recorded metric for SLO: {slo_name}",
operation="record_metric",
value=value,
success=success
)
async def _check_slo_compliance(self, slo_name: str):
"""Check SLO compliance and trigger alerts if needed."""
if slo_name not in self.slo_targets:
return
slo_target = self.slo_targets[slo_name]
current_value = await self.calculate_current_slo(slo_name)
if current_value is None:
return
# Determine violation severity
violation_severity = None
if current_value < slo_target.get_critical_value():
violation_severity = AlertSeverity.CRITICAL
elif current_value < slo_target.get_warning_value():
violation_severity = AlertSeverity.WARNING
# Handle violation
if violation_severity:
await self._handle_violation(slo_name, current_value, violation_severity)
else:
# No violation, clear any existing violation
await self._clear_violation(slo_name)
async def _handle_violation(
self,
slo_name: str,
actual_value: float,
severity: AlertSeverity
):
"""Handle SLO violation."""
slo_target = self.slo_targets[slo_name]
now = datetime.utcnow()
# Check if this is a new violation or continuation
if slo_name in self.current_violations:
# Update existing violation
violation = self.current_violations[slo_name]
violation.duration_minutes = (now - violation.violation_time).total_seconds() / 60
violation.actual_value = actual_value
violation.severity = severity
else:
# New violation
violation = SLOViolation(
slo_name=slo_name,
violation_time=now,
actual_value=actual_value,
target_value=slo_target.target_value,
severity=severity,
duration_minutes=0.0,
details={
"slo_type": slo_target.slo_type.value,
"time_window": slo_target.time_window.value,
"description": slo_target.description
}
)
self.current_violations[slo_name] = violation
self.violations[slo_name].append(violation)
self.stats["total_violations"] += 1
# Send alerts
await self._send_alert(violation)
self.logger.warning(
f"SLO violation: {slo_name}",
operation="slo_violation",
actual_value=actual_value,
target_value=slo_target.target_value,
severity=severity.value,
duration_minutes=violation.duration_minutes
)
async def _clear_violation(self, slo_name: str):
"""Clear SLO violation when back in compliance."""
if slo_name in self.current_violations:
violation = self.current_violations.pop(slo_name)
self.logger.info(
f"SLO violation cleared: {slo_name}",
operation="slo_violation_cleared",
total_duration_minutes=violation.duration_minutes
)
async def _send_alert(self, violation: SLOViolation):
"""Send alert for SLO violation."""
for callback in self.alert_callbacks:
try:
if asyncio.iscoroutinefunction(callback):
await callback(violation)
else:
callback(violation)
self.stats["alerts_sent"] += 1
except Exception as e:
self.logger.error(
f"Failed to send alert for {violation.slo_name}",
operation="send_alert",
error=e
)
async def calculate_current_slo(self, slo_name: str) -> Optional[float]:
"""Calculate current SLO value."""
if slo_name not in self.slo_targets:
return None
slo_target = self.slo_targets[slo_name]
metrics = self._get_metrics_for_window(slo_name, slo_target.time_window)
if not metrics:
return None
# Calculate based on SLO type
if slo_target.slo_type == SLOType.AVAILABILITY:
return self.calculator.calculate_availability(metrics)
elif slo_target.slo_type == SLOType.LATENCY:
# For latency SLOs, we want percentage under threshold
under_threshold = sum(1 for m in metrics if m.value <= slo_target.target_value)
return (under_threshold / len(metrics)) * 100
elif slo_target.slo_type == SLOType.ERROR_RATE:
error_rate = self.calculator.calculate_error_rate(metrics)
# For error rate SLOs, we want percentage compliance (low error rate)
return max(0, 100 - error_rate)
elif slo_target.slo_type == SLOType.THROUGHPUT:
time_window_minutes = self._get_time_window_minutes(slo_target.time_window)
throughput = self.calculator.calculate_throughput(metrics, time_window_minutes)
# Calculate percentage of target throughput achieved
return min(100, (throughput / slo_target.target_value) * 100)
elif slo_target.slo_type == SLOType.CUSTOM:
# For custom SLOs, use the metric values directly
if metrics:
return statistics.mean(m.value for m in metrics)
return None
def _get_metrics_for_window(self, slo_name: str, time_window: TimeWindow) -> List[SLOMetric]:
"""Get metrics for the specified time window."""
if slo_name not in self.metrics:
return []
now = datetime.utcnow()
window_minutes = self._get_time_window_minutes(time_window)
cutoff_time = now - timedelta(minutes=window_minutes)
return [
metric for metric in self.metrics[slo_name]
if metric.timestamp > cutoff_time
]
def _get_time_window_minutes(self, time_window: TimeWindow) -> float:
"""Convert time window to minutes."""
window_map = {
TimeWindow.MINUTE_1: 1,
TimeWindow.MINUTE_5: 5,
TimeWindow.MINUTE_15: 15,
TimeWindow.HOUR_1: 60,
TimeWindow.HOUR_4: 240,
TimeWindow.HOUR_24: 1440,
TimeWindow.DAY_7: 10080,
TimeWindow.DAY_30: 43200
}
return window_map.get(time_window, 60)
def get_slo_status(self, slo_name: str) -> Dict[str, Any]:
"""Get current status of an SLO."""
if slo_name not in self.slo_targets:
return {"error": f"Unknown SLO: {slo_name}"}
slo_target = self.slo_targets[slo_name]
current_value = asyncio.run(self.calculate_current_slo(slo_name))
# Calculate error budget
if current_value is not None:
error_budget_consumed = max(0, (slo_target.target_value - current_value) / slo_target.target_value * 100)
else:
error_budget_consumed = 0
# Get violation history
recent_violations = [
{
"timestamp": v.violation_time.isoformat(),
"severity": v.severity.value,
"duration_minutes": v.duration_minutes,
"actual_value": v.actual_value
}
for v in self.violations[slo_name][-10:] # Last 10 violations
]
return {
"slo_name": slo_name,
"target": slo_target.target_value,
"current_value": current_value,
"compliance_percentage": (current_value / slo_target.target_value * 100) if current_value else 0,
"error_budget_consumed_percentage": error_budget_consumed,
"status": "compliant" if slo_name not in self.current_violations else "violated",
"time_window": slo_target.time_window.value,
"description": slo_target.description,
"current_violation": (
{
"severity": self.current_violations[slo_name].severity.value,
"duration_minutes": self.current_violations[slo_name].duration_minutes,
"started_at": self.current_violations[slo_name].violation_time.isoformat()
}
if slo_name in self.current_violations else None
),
"recent_violations": recent_violations,
"total_metrics": len(self.metrics[slo_name])
}
def get_all_slo_status(self) -> Dict[str, Any]:
"""Get status of all SLOs."""
slo_statuses = {}
for slo_name in self.slo_targets.keys():
slo_statuses[slo_name] = self.get_slo_status(slo_name)
# Calculate overall compliance
compliant_slos = sum(
1 for status in slo_statuses.values()
if status.get("status") == "compliant"
)
total_slos = len(slo_statuses)
overall_compliance = (compliant_slos / total_slos * 100) if total_slos > 0 else 100
return {
"timestamp": datetime.utcnow().isoformat(),
"overall_compliance_percentage": overall_compliance,
"total_slos": total_slos,
"compliant_slos": compliant_slos,
"violated_slos": total_slos - compliant_slos,
"statistics": self.stats,
"slo_details": slo_statuses
}
def get_error_budget_report(self) -> Dict[str, Any]:
"""Get error budget consumption report."""
report = {
"timestamp": datetime.utcnow().isoformat(),
"error_budgets": {}
}
for slo_name, slo_target in self.slo_targets.items():
current_value = asyncio.run(self.calculate_current_slo(slo_name))
if current_value is not None:
# Calculate error budget
budget_remaining = max(0, current_value - slo_target.target_value)
budget_consumed = max(0, slo_target.target_value - current_value)
budget_total = 100 - slo_target.target_value # Assuming percentage-based
consumption_percentage = (budget_consumed / budget_total * 100) if budget_total > 0 else 0
report["error_budgets"][slo_name] = {
"target": slo_target.target_value,
"current_value": current_value,
"budget_consumed": budget_consumed,
"budget_remaining": budget_remaining,
"consumption_percentage": consumption_percentage,
"status": (
"healthy" if consumption_percentage < 50
else "warning" if consumption_percentage < 80
else "critical"
),
"time_window": slo_target.time_window.value
}
return report
# Global SLO monitor
slo_monitor = SLOMonitor()
# Convenience functions for common metrics
def record_api_request(endpoint: str, response_time_ms: float, success: bool):
"""Record API request metrics for SLO monitoring."""
slo_monitor.record_metric("api_availability", 1.0, success, {"endpoint": endpoint})
slo_monitor.record_metric("api_latency_p95", response_time_ms, success, {"endpoint": endpoint})
def record_investigation_result(investigation_id: str, success: bool, duration_ms: float):
"""Record investigation result for SLO monitoring."""
slo_monitor.record_metric("investigation_success_rate", 1.0, success, {
"investigation_id": investigation_id,
"duration_ms": duration_ms
})
def record_agent_task(agent_name: str, task_type: str, success: bool, duration_ms: float):
"""Record agent task for SLO monitoring."""
slo_monitor.record_metric("agent_error_rate", 1.0, success, {
"agent_name": agent_name,
"task_type": task_type,
"duration_ms": duration_ms
})
def record_database_query(query_type: str, duration_ms: float, success: bool):
"""Record database query for SLO monitoring."""
slo_monitor.record_metric("database_latency_p90", duration_ms, success, {
"query_type": query_type
})
def record_anomaly_detection(true_positive: bool, false_positive: bool):
"""Record anomaly detection accuracy."""
if true_positive:
accuracy = 100.0
elif false_positive:
accuracy = 0.0
else:
accuracy = 50.0 # Unknown case
slo_monitor.record_metric("anomaly_detection_accuracy", accuracy, True, {
"true_positive": true_positive,
"false_positive": false_positive
})
# Default alert callback
async def default_alert_callback(violation: SLOViolation):
"""Default alert callback that logs violations."""
logger.warning(
f"SLO VIOLATION ALERT: {violation.slo_name}",
operation="slo_alert",
severity=violation.severity.value,
actual_value=violation.actual_value,
target_value=violation.target_value,
duration_minutes=violation.duration_minutes
)
# Record alert metric
BusinessMetrics.record_investigation_created() # Using as example metric
# Register default alert callback
slo_monitor.register_alert_callback(default_alert_callback) |