Cloud Events Reader Examples¶
Cloud Events Reader is a component that collects Kubernetes events and forwards them to the logging pipeline. These examples show different deployment scenarios for various operational requirements.
Simple Deployment¶
Basic Cloud Events Reader configuration suitable for most environments:
This configuration provides:
- Basic resource allocation (100m CPU, 128Mi memory)
- Standard event collection from Kubernetes API
- Minimal resource footprint
Node Selector Configuration¶
Targeted deployment on specific nodes using node selectors:
This configuration adds:
- Node selector for targeted deployment
- Same resource allocation as simple deployment
- Useful for dedicated logging nodes or specific node pools
Key Configuration Parameters¶
| Parameter | Description | Default |
|---|---|---|
cloudEventsReader.install |
Enable/disable Cloud Events Reader deployment | false |
cloudEventsReader.resources |
Resource requests and limits | - |
cloudEventsReader.nodeSelector |
Node selection criteria | - |
Container hardening¶
Cloud Events Reader runs with UID and GID 1000 on Kubernetes. On OpenShift, the platform assigns a non-root UID while
the workload retains primary GID 1000. The container uses a read-only root filesystem, disables privilege escalation,
drops all Linux capabilities, and uses the RuntimeDefault seccomp profile. A 100 MiB emptyDir mounted at /tmp
provides its only writable runtime path.
The service account can only get, list, and watch Kubernetes Events through the core/v1 and events.k8s.io APIs. It
does not receive the built-in view role and cannot read pods, ConfigMaps, or Secrets.
Use Cases¶
- Simple Deployment: Standard Kubernetes clusters with default scheduling
- Node Selector: Clusters with dedicated nodes for logging components
- Resource Constraints: Environments requiring specific resource allocation