Seiri guides
Kubernetes CronJob monitoring
Kubernetes CronJob monitoring needs more than kubectl. Pods get evicted mid-run; jobs show no error but never reach completion. Seiri covers both: ping on success, or install the HealthCheck operator.
Cron job monitoring
Heartbeat monitoring
Dead man's switch
Scheduled jobs
K8s CronJobs
Kube agent
Email heartbeats
Background jobs
Two approaches
1) Add curl to ping.seiri.app/ as the last step of the CronJob container.
2) Install seiri-kube-agent (Helm) and create HealthCheck CRs for Jobs/Deployments — phases like Healthy, Unhealthy, Missing stream to Seiri.
Helm install
helm install seiri-kube-agent ./helm-main \
-n seiri-system --create-namespace \
--set seiri.authToken="…" \
--set seiri.clusterId="…"
Docs
Full chart reference: seiri-app.github.io/seiri-kube-agent. Product guide: /kubernetes-agent.
FAQ
Is seiri-kube-agent only for CronJobs?
No. HealthChecks can target Deployments, Pods, Jobs, StatefulSets, and DaemonSets.
Do I need both ping and the agent?
Either works. Many teams ping CronJobs and use the operator for long-running Deployments.