As per https://github.com/openshift/cluster-logging-operator/pull/1519, spec.collection.logs has been deprecated and replaced with spec.collection. Yet documentation for example in https://docs.openshift.com/container-platform/4.13/logging/config/cluster-logging-collector.html still shows and uses spec.collection.logs which should be fixed to get customers to move to the new API specification.

OpenShift Container Platform 4.13 and older

Always

1. Check https://docs.openshift.com/container-platform/4.13/logging/config/cluster-logging-collector.html#cluster-logging-collector-limits_cluster-logging-collector
2. Example shown is using spec.collection.logs

Example showing:
spec:
  collection:
    logs:
      fluentd:
        resources:
          limits: 
            memory: 736Mi
          requests:
            cpu: 100m
            memory: 736Mi

Example should show
spec:
  collection:
    fluentd:
      resources:
        limits: 
          memory: 736Mi
        requests:
          cpu: 100m
          memory: 736Mi

Eventually more examples, etc. need to be updated but we should avoid examples in general with deprecated API specification