https://github.com/openshift/cluster-authentication-operator/pull/587 addresses an issue in which the auth operator goes degraded when the console capability is not enabled.  The rest is that the console publicAssetURL is not configured when the console is disabled.  However if the console capability is later enabled on the cluster, there is no logic in place to ensure the auth operator detects this and performs the configuration.

Manually restarting the auth operator will address this, but we should have a solution that handles it automatically.

4.12

Always

1. Install a cluster w/o the console cap
2. Inspect the auth configmap, see that assetPublicURL is empty
3. Enable the console capability, wait for console to start up
4. Inspect the auth configmap and see it is still empty

assetPublicURL does not get populated

assetPublicURL is populated once the console is enabled