Multiple VMs Fails to Boot after restoring them to new cluster (but not all)

[helojunkie]

Hello - I have a very weird issue I am trying to solve. I had a stand-alone promox server that has been running fantastic for quite some time. We decided to move to a more powerful proxmox custer. I utilize shard storage for all of our vzdump backups. After getting the new cluster up and running I restored all of my VMs to the new cluster. Everything went smoothly and all VM cam up with no problems.

This morning when we came into the office, several VMs had failed with a message that they had no boot device. This is AFTER we restored them, started them and tested them.

When I restored them (AGAIN) they started up just fine, but after a few hours, they fail with the exact same message, that they did not have a boot device. I am at a loss as to where to even start looking for this problem. Why would they restore, startup just fine, and then fail some hours later with no boot device?

To be clear, this is happening to only three VMs out of 30 on the cluster and it is across nodes and across operating systems (2 Linux, 1 Windows).

Where would I go to troubleshoot this issue?

EDIT: Also, not sure why but a bunch of my Linux VMs are reporting READ ONLY file system when attempting to do anything with them.

Many Thanks

 

[oguz]

hi,

To be clear, this is happening to only three VMs out of 30 on the cluster and it is across nodes and across operating systems (2 Linux, 1 Windows).

* can you please post the VM configurations of these? qm config VMID

* and also post the output of pveversion -v from your nodes (if they are all the same package versions its enough to post from one)

* did you enable High Availability by any chance?