[BETA] HPE VME – Backup Job Stuck in “Stopping” State

[marco.mosca]

Hello,

A backup job became stuck at the “Preparing for backup” stage immediately after detecting the availability of a Worker.

After waiting for some time without any progress, we attempted to execute the Stop command.

From that moment on, the job entered the “Stopping” state and can no longer be managed.
Attempting to run Stop again from the console returns the following error:
“Failed to stop the backup job xxxxx. It is not running.”

Any suggestions on how to resolve this issue?

Best regards
Marco

[cody.ault]

Hey Marco,

I'll reach out to you directly to collect some log files. Did you create the worker on the Management OVS bridge or did you put it on an OVS Port Group for VMs?

[EvgenyBaev]

If the worker has already been created, the issue is most likely related to a failure during the initial libvirt connection.

You can find more details in the following forum post:
post562086.html#p562086

It would be very helpful for us if you could review that post and provide answers to the questions mentioned there.

I have a few follow-up questions for everyone involved:
1) Could you please check if port 7443 is open on your HPE host?

Code: Select all

sudo iptables -L INPUT -n --line-numbers

2) Does this issue occur consistently in your lab, or is it intermittent? If it is intermittent, there may be an issue with the connection from worker side. In this case, the planned retry logic should help. We can provide a hotfix for testing.
3) Was this a completely new deployment of the HPE server with version 8.0.11/8.0.12, or did you upgrade your lab from an earlier build?

This information will assist us in troubleshooting the issue further.

As for this one.

From that moment on, the job entered the “Stopping” state and can no longer be managed.
Attempting to run Stop again from the console returns the following error:
“Failed to stop the backup job xxxxx. It is not running.”

One possible way to fix this issue is through the VBR database. I’ll first check if there’s a simpler solution. If not, I’ll provide instructions on how to resolve it via the database.