Files
homelab-infra/platform/observability/metrics/rules/monitoring-delivery.yaml
T
panxiao81 39aeac25ee
yaml / yaml (push) Successful in 1m10s
补齐集群状态、主机与通知链路基础监控 (#149)
Co-authored-by: panxiao81 <[email protected]>
2026-09-25 17:45:56 +00:00

59 lines
2.7 KiB
YAML

apiVersion: operator.victoriametrics.com/v1beta1
kind: VMRule
metadata:
name: monitoring-delivery
namespace: monitoring
labels:
app.kubernetes.io/part-of: victoria-metrics
spec:
groups:
- name: monitoring-delivery
interval: 30s
rules:
- alert: KubeStateMetricsUnavailable
expr: up{job="kube-state-metrics"} == 0 or absent(up{job="kube-state-metrics"})
for: 5m
labels:
severity: critical
annotations:
summary: kube-state-metrics 采集不可用
description: 检查 HelmRelease、ServiceMonitor 转换、targets 与 exporter;目标消失也会触发。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-foundation.md
- alert: NodeExporterUnavailable
expr: up{job="node-exporter"} == 0 or absent(up{job="node-exporter"})
for: 5m
labels:
severity: critical
annotations:
summary: node-exporter 采集不可用
description: 检查节点与 exporter,主机健康规则依赖此采集。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-foundation.md
- alert: NatsMetricsUnavailable
expr: up{job="nats/nats"} == 0 or absent(up{job="nats/nats"})
for: 5m
labels:
severity: warning
annotations:
summary: NATS 指标采集不可用
description: 依次检查 PodMonitor、转换后的 VMPodScrape、:7777 exporter 及 targets。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-foundation.md
- alert: AlertmanagerTelegramDeliveryFailed
expr: sum by(job,instance,integration) (increase(alertmanager_notifications_failed_total{job="vmalertmanager-main",integration="telegram"}[5m]))
> 0
for: 1m
labels:
severity: critical
annotations:
summary: Alertmanager 向 Telegram 发送失败
description: 检查网络、bot token 和频道权限。若 Telegram 完全不可达,本告警也无法经同一渠道送达;需在 Alertmanager/Grafana 排查,集群外心跳另行建设。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-foundation.md
- alert: AlertmanagerConfigurationReloadFailed
expr: alertmanager_config_last_reload_successful{job="vmalertmanager-main"} == 0
for: 5m
labels:
severity: critical
annotations:
summary: Alertmanager 配置加载失败
description: 检查 operator 和 Alertmanager 日志、配置格式及 Secret 引用。
runbook_url: https://git.ddupan.top/panxiao81/homelab-wiki/src/branch/main/guides/monitoring-foundation.md