I have this error on my resource group listing which happened after a power failure. I am not sure what it means or how to fix it. Can someone point me in the right direction?
What is the output of linstor node list on the controller?
It looks like sv2 is not connected to any of its peers, which could be due to a lingering network issue, but the node listing should tell us more.
I’d suggest you try to restart the linstor-satellite service on each of your nodes, and then rerun your resource-group listing on the controller to see if anything changes.
If its still showing in that Connecting state, you may want to double check the satellite nodes are still able to reach srv2 via the currently configured IP address and port for those resources, especially if you have a dedicated replication network that is different than your management network.
Otherwise the kernel logs of those satellites may have some clues when you review the DRBD log lines for those resources between the peers.
Hi,
I did restart the linstor-satellite on each node but had no success. Connectivity between nodes was also good. I realized that the disk that was in error belonged to a LXC that was part of a HA (high avail) group in proxmox. Since I had a good backup, I restored the LXC and deleted the original LXC and the disk of course was also deleted. That cleared up the problem.
I think that the HA interaction along with the power failure caused the problem. Could be the LXC was trying to be restarted on an another node in the HA group after the power went. The servers are on UPS but they turned off several mins apart after power failure.
Thank you for your help
Looks like you have to (dis)connect the drbd resource with its peers.e.g.: drbdadm disconnect pm-71d8786e && drbdadm connect pm-71d8786e
That may still not fully resolve the issue, but then you need to show drbd looks, to decide if --my-disard-my-data or you need fresh metadata on a device.