T8.411.8.2026תיאום: BAROUCH SHALITReview v1.0
Observability, Alerts ו-SLO
לזהות כשל לפני שהוא הופך לפגיעה עסקית — הפרדת קטגוריות, SLOs, Alert Rules עם owner/runbook, health checks, false positive/negative.
קריטריוני קבלה
- לכל P0/P1 יש התראה
- אין alert ללא owner
- False positive/negative נבדקו
הפרדת קטגוריות (4)
Business
ביצוע עסקאות, P&L, מזומנים, עמדות
Infrastructure
APIs, DB, Workflows, זמינות, latency
Security
אימות, הרשאות, חריגות, Break-Glass, Audit
Data Quality
טריות נתונים, סתירות, חסרים, Quarantine
SLOs (6)
| ID | שם | שירות | מדד | יעד | Error Budget | קטגוריה | Owner |
|---|---|---|---|---|---|---|---|
| SLO-001 | Trade Execution Success Rate | trade_execution | success_rate | 99.5% | 0.5% | business | BAROUCH SHALIT |
| SLO-002 | Market Data Freshness | market_data | data_freshness | 95% | 5% | data_quality | BAROUCH SHALIT |
| SLO-003 | Agent Scan Availability | agent_scan | availability | 99% | 1% | infrastructure | BAROUCH SHALIT |
| SLO-004 | Portfolio Refresh Latency P99 | portfolio_refresh | latency_p99 | 5000ms | 1000ms | infrastructure | BAROUCH SHALIT |
| SLO-005 | FX Rate Update Success | fx_rate | success_rate | 99% | 1% | infrastructure | BAROUCH SHALIT |
| SLO-006 | Audit Hash Chain Integrity | audit_integrity | success_rate | 100% | 0% | security | BAROUCH SHALIT |
Alert Rules (10)
| ID | שם | חומרה | קטגוריה | תנאי | Owner | FP | FN |
|---|---|---|---|---|---|---|---|
| ALR-001 | Trade Execution Failed | P0 | business | trade.status=FAILED | BAROUCH SHALIT | ||
| ALR-002 | Global Freeze Activated | P0 | security | settings.global_freeze=true | BAROUCH SHALIT | ||
| ALR-003 | Break-Glass Event | P0 | security | BreakGlassEvent.created | BAROUCH SHALIT | ||
| ALR-004 | Audit Hash Chain Broken | P0 | security | verifyHashChain.valid=false | BAROUCH SHALIT | ||
| ALR-005 | Market Data Stale (>15min) | P1 | data_quality | data_freshness>15min | BAROUCH SHALIT | ||
| ALR-006 | Daily Loss Limit Breached | P1 | business | daily_loss>limit_pct | BAROUCH SHALIT | ||
| ALR-007 | Agent Scan Failed | P1 | infrastructure | agent_scan.error | BAROUCH SHALIT | ||
| ALR-008 | FX Rate Fetch Failed | P2 | infrastructure | fx_rate.error | BAROUCH SHALIT | ||
| ALR-009 | Portfolio Price Lag | P2 | data_quality | price_lag>5min | BAROUCH SHALIT | ||
| ALR-010 | Data Quality Conflict | P2 | data_quality | DataQualityLog.flag=CONFLICTED | BAROUCH SHALIT |
הרצת בדיקות T8.4
לחץ "הרץ בדיקות" להפעלת 9 בדיקות Observability (SLO coverage, alert coverage, owner/runbook, FP/FN, health check).
ראיות
- SLO entity: 6 SLOs עם target, error_budget, owner, runbook, category
- AlertRule entity: 10 כללים — 4 P0 + 3 P1, כולם עם owner + runbook
- Health checks: SystemSettings, AgentConfig, AuditEvent, Incident — כולם reachable
- FP/FN: כל P0/P1 נבדקו False Positive + False Negative
- Incident entity: קיימת עם timeline + post-incident review + correlation_id + break_glass_event_id
InvestIQ · פרק 8 · T8.4 Observability, Alerts ו-SLO · Review v1.0 · 11.8.2026 · תיאום: BAROUCH SHALIT