k8s-event-logger
Wrapper around the upstream k8s-event-logger Helm chart (v1.1.8). Watches the Kubernetes API for Events and writes them to its own stdout as structured log lines, so they can be picked up by whichever log stack the cluster runs.
Why it's in KubeAid
Kubernetes Events (kubectl get events) are ephemeral and namespaced by default - once they expire from
etcd they're gone. This chart turns them into durable, centrally-searchable log lines by shipping them
through the cluster's log pipeline (Graylog, OpenObserve, or OpenSearch) alongside application logs.
Key values / KubeAid-specific configuration
| Value | Description | Default |
|---|---|---|
networkpolicies | Enable this chart's NetworkPolicy resources | false |
k8s-event-logger.env.KUBERNETES_API_URL | API server URL the logger watches | https://kubernetes.default.svc |
All other configuration is forwarded to the upstream chart under the k8s-event-logger key; see the
vendored chart README.
Operational notes
- When
networkpolicies: true, a CalicoNetworkPolicy(templates/netpol-k8s-event-logger.yaml) allows egress to nodes labelledkubernetes.io/role == 'master'on port 443, so the logger can reach the apiserver even when a default-deny egress policy is in place.
Docs links
- Upstream chart README:
charts/k8s-event-logger/README.md - Monitoring / log stack comparison