I recommend opening a service request for this issue. Troubleshooting this issue will require analyzing the MCS logs and they are not the most friendly logs.

I appreciate the suggestion, Ian, but opening a service request is not an option for us as we didn't renew maintenance knowing this system was being retired.

Having said that, I did come up with an unorthodox work-around that involved interacting with the PostgreSQL database and restarting the MCS.  Yeah, I know, not exactly a supported fix.

The goal is to power down the Avamar and either sell it or scrap it.  I'm leaning towards the later.  We ceased all backup activity four months ago and retired all clients at that time.  Since then, all backups have been aging off the system naturally. With all backups now gone, we've been running manual garbage collection in order to purge the last remaining data.  The only detail left was deleting the now-retired clients.

I'm not certain what the goal with deleting these clients is but removing them from the MCS database won't actually delete them in a real sense -- the clients and any associated data that has not yet expired will still exist in the GSAN.

If the idea is just to remove the clients from the GUI, hacking up the mcdb will do what you're after but if you're trying to purge the data, it will not.

Are you facing this issue while deleting the VM client or the physical guest machine.?

Only with the VM clients.

I faced the same issue in our environment.

Reason is, the vcenter is retired and the VM's became orphaned and unable to delete as the Vcenter services/connections on avamar couldn't contact the Vcenter.

Raised a SR with EMC, support engineer ran some scripts and added the entry of the Vcenter client on the Vcenter domain which was missing to the MCS database.(you should see the vcenter server client entry with ? icon on the vcenter domain). post that I was able to delete the VM.

This helped me indirectly.

I couldn't RETIRE one of VM clients (all others retired without issues, backups and restores all worked normally).

After reding your comment I checked the vCenter cloient in vCenter domain and it wasn't activated. I activated it and after that the client retired normally.

Happy to resolve the issue but it leaves me with a big question, Why only this one client?