Files
MoonBridge/deployment/observability/prometheus/strataproxy-alerts.yml
T

115 lines
4.3 KiB
YAML

groups:
- name: strataproxy.availability
rules:
- alert: StrataProxyTargetDown
expr: up{job="strataproxy"} == 0
for: 1m
labels:
severity: critical
annotations:
summary: "StrataProxy scrape target is down"
description: "Prometheus cannot scrape the StrataProxy Admin API metrics endpoint."
- alert: StrataProxyNoReadyBackends
expr: strataproxy_servers_registered == 0 or sum(strataproxy_server_health{status="UP"}) == 0
for: 2m
labels:
severity: critical
annotations:
summary: "StrataProxy has no ready backend"
description: "No registered backend is currently healthy enough to receive players."
- alert: StrataProxyRouteFailures
expr: increase(strataproxy_routes_failed_total[5m]) > 0
for: 2m
labels:
severity: warning
annotations:
summary: "StrataProxy route failures detected"
description: "The proxy failed to route at least one player in the last five minutes."
- alert: StrataProxyBackendConnectFailures
expr: increase(strataproxy_backend_connect_failures_total[5m]) > 5
for: 2m
labels:
severity: warning
annotations:
summary: "StrataProxy backend connection failures"
description: "Backend connection failures are above the expected baseline."
- name: strataproxy.performance
rules:
- alert: StrataProxyEventLoopDelayHigh
expr: strataproxy_event_loop_delay_seconds > 0.025
for: 2m
labels:
severity: warning
annotations:
summary: "StrataProxy event loop delay is high"
description: "Observed event loop delay is above 25ms. Check CPU saturation, blocking work, and compression rewrite policy."
- alert: StrataProxyEventLoopDelayCritical
expr: strataproxy_event_loop_delay_seconds > 0.100
for: 1m
labels:
severity: critical
annotations:
summary: "StrataProxy event loop delay is critical"
description: "Observed event loop delay is above 100ms and player forwarding latency is likely affected."
- alert: StrataProxyHeapPressure
expr: strataproxy_jvm_heap_used_bytes / strataproxy_jvm_heap_max_bytes > 0.85
for: 5m
labels:
severity: warning
annotations:
summary: "StrataProxy JVM heap pressure"
description: "Heap usage is above 85% of max heap for five minutes."
- alert: StrataProxyDirectMemoryPressure
expr: strataproxy_pooled_direct_memory_bytes > 1610612736
for: 5m
labels:
severity: warning
annotations:
summary: "StrataProxy pooled direct memory is high"
description: "Pooled direct memory is above 1.5GiB. Compare with -XX:MaxDirectMemorySize and active connection load."
- alert: StrataProxyRelayBackpressure
expr: increase(strataproxy_relay_backpressure_events_total[5m]) > 0
for: 5m
labels:
severity: warning
annotations:
summary: "StrataProxy relay backpressure"
description: "One or more relay directions observed non-writable target channels."
- name: strataproxy.security
rules:
- alert: StrataProxyRejectedConnectionSpike
expr: increase(strataproxy_connections_rejected_total[5m]) > 100
for: 2m
labels:
severity: warning
annotations:
summary: "StrataProxy rejected connection spike"
description: "Connection admission is rejecting a high number of clients; check storms and per-address limits."
- alert: StrataProxyPacketAnomalies
expr: increase(strataproxy_packet_anomalies_total[5m]) > 0
for: 2m
labels:
severity: warning
annotations:
summary: "StrataProxy packet anomalies detected"
description: "Packet anomaly rules are firing. Inspect /packet-anomalies or strataproxy-admin anomalies."
- alert: StrataProxyCustomPayloadFlood
expr: increase(strataproxy_packet_anomalies_total{rule="custom-payload-flood"}[5m]) > 0
for: 1m
labels:
severity: warning
annotations:
summary: "StrataProxy custom payload flood"
description: "A player connection triggered the custom payload flood rule."