# Observability & Monitoring Strategy ## Prometheus & Grafana Metric Targets --- ## Core Business & Operational Metrics | Metric Name | Target Threshold | Alerting Condition | | :--- | :---: | :--- | | `http_requests_total` | N/A | High 5xx error rate (>1% over 5m) | | `p95_search_latency_ms` | $< 80\text{ ms}$ | Latency $> 200\text{ ms}$ over 5m | | `p95_checkout_latency_ms` | $< 200\text{ ms}$ | Latency $> 500\text{ ms}$ over 5m | | `payment_success_rate` | $> 98\%$ | Drop below 90% over 10m | | `pos_sync_queue_depth` | $< 50$ items | Queue depth $> 500$ over 15m | | `opensearch_circuit_breaker_tripped` | `0` | Tripped == `1` (Immediate PagerDuty) | | `mysql_replication_lag_seconds` | $< 2\text{ sec}$ | Lag $> 10\text{ sec}$ |