Alert pack

KubeMeridian ships a curated Prometheus Operator PrometheusRule with 25 alerts across the common Kubernetes failure modes, evaluated by your own Prometheus / Alertmanager.

Install

kubectl apply -f deploy/kubemeridian-prometheusrule.yaml

Set the release: label to match your Prometheus Operator ruleSelector (kube-prometheus-stack uses the Helm release name).

Groups

Group	Alerts
nodes	NotReady, Memory/Disk/PID pressure, filesystem almost full
workloads	CrashLooping, OOMKilled, high restart rate, container waiting, Pending, Failed, Deployment/StatefulSet mismatch, DaemonSet rollout, Job failed, HPA maxed
resources	CPU throttling, quota almost full
storage	PV/PVC & inodes filling up (crit + warn), PV errors
monitoring-meta	KubeStateMetricsDown (highest-value), NodeExporterDown, TargetDown

TIP

kube-prometheus-stack already ships most of the canonical kubernetes-mixin alerts. If you run it with defaults, install only the bold "KubeMeridian addition" rules to avoid duplicates. For a raw Prometheus + KSM setup, apply the whole file. Full guidance is in the file header.