proxy-pool/deploy/prometheus/rules/proxy-pool.yml
2026-08-07 17:14:54 +08:00

75 lines
2.8 KiB
YAML

groups:
- name: proxy-pool
rules:
- alert: ProxyPoolGatewayOutcomeFailureRate
expr: |
sum(rate(proxy_pool_gateway_outcomes_total{result="failure"}[5m]))
/ clamp_min(sum(rate(proxy_pool_gateway_outcomes_total[5m])), 1) > 0.1
for: 10m
labels:
severity: warning
annotations:
summary: Gateway 代理尝试失败比例持续高于 10%
- alert: ProxyPoolGatewayHighLatency
expr: |
histogram_quantile(0.99,
sum by (le, protocol) (rate(proxy_pool_gateway_request_duration_seconds_bucket[5m]))) > 5
for: 10m
labels:
severity: warning
annotations:
summary: Gateway p99 请求时延持续高于 5 秒
- alert: ProxyPoolGatewayOutcomeQueueDrops
expr: sum(increase(proxy_pool_gateway_outcome_queue_dropped_total[5m])) > 0
for: 5m
labels:
severity: warning
annotations:
summary: Gateway Outcome 本地队列发生丢弃
- alert: ProxyPoolNoAvailableSlots
expr: |
sum(proxy_pool_controller_capacity_active_upstreams) > 0
and sum(proxy_pool_controller_capacity_available_slots) == 0
for: 1m
labels:
severity: critical
annotations:
summary: Controller 可分配容量耗尽
- alert: ProxyPoolCapacityInventoryErrors
expr: sum(rate(proxy_pool_controller_capacity_inventory_reads_total{result="error"}[10m])) > 0.2
for: 10m
labels:
severity: warning
annotations:
summary: Controller 容量库存读取持续失败
- alert: ProxyPoolProviderFetchErrors
expr: sum(rate(proxy_pool_controller_provider_fetch_results_total{class="error"}[10m])) > 0.2
for: 10m
labels:
severity: warning
annotations:
summary: Provider 获取持续失败
- alert: ProxyPoolCheckerObservationRejections
expr: |
sum(rate(proxy_pool_checker_observations_total{result="rejected"}[5m]))
/ clamp_min(sum(rate(proxy_pool_checker_observations_total[5m])), 1) > 0.05
for: 10m
labels:
severity: warning
annotations:
summary: Checker Observation 拒绝比例持续高于 5%
- alert: ProxyPoolExtractionIdempotencyConflict
expr: sum(rate(proxy_pool_controller_extraction_requests_total{result="idempotency_conflict"}[5m])) > 0
for: 5m
labels:
severity: warning
annotations:
summary: Distribution 幂等键冲突持续发生
- alert: ProxyPoolControlPlaneTargetDown
expr: up{job=~"proxy-gateway|proxy-controller|proxy-checker"} == 0
for: 2m
labels:
severity: critical
annotations:
summary: Proxy Pool 控制面目标不可抓取