Skip to main content

seiri-kube-agent · Helm chart

Kubernetes HealthCheck operator for Seiri

Deploy a lightweight operator that watches HealthCheck CRs and continuously monitors Deployments, Pods, StatefulSets, DaemonSets, and Jobs — then reports results to cloud.seiri.app for dashboards and alerts.

Register a cluster Helm chart & install

What gets installed

  • · CRD healthchecks.monitoring.seiri.app (monitoring.seiri.app/v1alpha1)
  • · Deployment running ghcr.io/seiri-app/kubeagent
  • · ServiceAccount, ClusterRole, ClusterRoleBinding (least privilege)
  • · Secret for your Seiri auth token

Helm Chart

Kubernetes operator that monitors cluster health and reports status to the Seiri observability platform.

Prerequisites: Kubernetes 1.21+, Helm 3.x, and a cluster registered in Seiri (for authToken + clusterId).

Install via Helm repo

helm · chart repository shell 
helm repo add seiri https://seiri-app.github.io/seiri-kube-agent
helm repo update
helm install seiri-agent seiri/seiri-kube-agent

Install via OCI (GHCR)

helm · OCI registry shell 
helm install seiri-agent oci://ghcr.io/seiri-app/charts/seiri-kube-agent

Wire it to your cluster

values from your Seiri cluster registration shell 
helm install seiri-agent seiri/seiri-kube-agent \
  --namespace seiri-system --create-namespace \
  --set seiri.authToken="<YOUR_AUTH_TOKEN>" \
  --set seiri.clusterId="<YOUR_CLUSTER_ID>"

Verify

kubectl shell 
kubectl get pods -n seiri-system
kubectl get crd healthchecks.monitoring.seiri.app
GitHub ↗ Chart Index ↗

Example HealthCheck

apiVersion: monitoring.seiri.app/v1alpha1
kind: HealthCheck
metadata:
  name: check-test
  namespace: default
spec:
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: my-app
  interval: 10s
  checkID: "chk-test-001"
  tags:
    env: development
    region: local

kubectl apply -f check-test.yaml then kubectl get healthchecks -A — columns include Phase, Last Check, Last Success, Last Failed.

Target kinds & phases

targetRef.kind

Deployment · Pod · Job · StatefulSet · DaemonSet

Phase

Healthy · Unhealthy · Missing · Error · Unknown · Suspended

Security defaults

Non-root UID 65532, read-only root filesystem, no privilege escalation, RuntimeDefault seccomp. Auth token lives in a Secret and is injected as an env var — never as a mounted file.

Ready to wire a cluster?

Create the cluster in Seiri, copy authToken + clusterId, install the chart, apply a HealthCheck.

Open Seiri Helm chart reference

FAQ

Does uninstall remove HealthChecks?

Helm uninstall removes the operator; CRD and HealthCheck resources are not deleted automatically. Clean up with kubectl delete healthchecks --all -A and delete the CRD if desired.

Can I still ping CronJobs over HTTP?

Yes. Heartbeat webhooks via ping.seiri.app/<ref> still work alongside in-cluster HealthChecks. See cron monitoring.