Goal

Add the openshift cluster ID  to log records.

The cluster ID is part of the `clusterversion` resource, can be retrieved with:

oc get clusterversion/version -o jsonpath='{.spec.clusterID}{"\n"}' 

Why is this important?

When logs from multiple clusters are aggregated in the same store, it must be possible to distinguish logs from different clusters.

Scenarios

1. ACM (Advanced Cluster Manager) may aggregate logs from multiple managed clusters.
2. RHOBS (Red Hat Observability Service)  manages multiple clusters on behalf of  multiple customers.
3. HyperShift may aggregate logs from many sub-clusters.
4. Telco customers may want to aggregate logs from many edge clusters in a single NOC (Network Operation Centre)

There are likely to be other scenarios.

Acceptance Criteria

• All log records, of all types (application, infrastructure, audit),  MUST have a new field `openshift.cluster_id` set to the value of the cluster ID.
• CI tests, technical enablement slides etc.
• When syslog.addLogSource is true, the cluster ID should be included with the other source information.

Notes

• Must be added to vector and fluentd

Docs

[rkratky] Not sure this needs explicit mention in docs. Perhaps only to update potential existing descriptions of log records.