Prometheus: Copy-Paste PromQL Queries for Kubernetes

rate(kube_pod_container_status_restarts_total[5m]) > 0
or
kube_pod_status_phase{phase!="Running", phase!="Succeeded"}

argocd_app_info{health_status!="Healthy"} > 0

argocd_app_info{sync_status="OutOfSync"} > 0

avg(node_load15) / count(node_cpu_seconds_total{mode="idle"}) > 0.8

avg(
  (node_memory_MemTotal_bytes - node_memory_MemFree_bytes
   - node_memory_Buffers_bytes - node_memory_Cached_bytes)
  / node_memory_MemTotal_bytes
) > 0.85

pods_crashing = {
  name            = "Pod Restarting Alert"
  no_data_state   = "OK"
  condition       = "rate(kube_pod_container_status_restarts_total[5m]) > 0"
  triggered_after = "5m"
  annotations = {
    severity = "critical"
    summary  = "Pod keeps restarting for 5+ minutes"
    todo     = "kubectl logs <pod> -n <ns> --previous"
  }
  labels = {
    team    = "devops"
    send_to = "teams_channel_devops"
  }
}