If a 0-db deployed as part of qsfs in zos goes down or is deleted, there is no easy way to figure that out. Ideally it would be replaced automatically and storage repaired. Since deployments require a mnemonic (thus highly sensitive data), we can't just put this is in the deployment itself.
Ideally there would be an active twin on the user side, which can do active health checks on the backends, but no such thing currently exists.