Node showing as offline after copying to it

[J4N11S]

I have a cluster with 3 nodes. All in all, this works perfectly. However, if I want to copy a VM or ISO from one of the first two nodes to the last node, the copy process aborts after about 1GB and the third node appears as offline. However, when I click on it, everything is displayed and updated in the dashboard correctly. After about 30 minutes everything works and it shows as online again.

I have tried all possible steps. Even if I restart all nodes, this node is still displayed as unavailable. When I open the web panel of the node, the other two nodes are shown as unreachable.

Have any of you ever had a similar scenario? I have also reinstalled all three nodes and still have the same problem.

 

[esi_y]

I would connect to the GUI of the "red" node and see for myself first, if it's alive and if it can see the other two nodes. To troubleshoot better, it would be nice to know pvecm status from the node that the migration fails from. If the cluster has full quorum, then it is either GUI glitch or you could check with e.g. journalctl --since "1 hour ago" -u corosync if you were losing quorum during the migration. If that's the case, it is highly likely your migration is on shared network with the cluster messaging and gets saturated, which disrupts the quorum, which stops cluster operations, which causes failed migration.