spmstatusvds fails and hosts servers are rebooted
Oracle Linux 8.8 with Unbreakable Enterprise Kernel [5.4.17]
Oracle Linux Virtualization Manager 4.4.10.7-1.0.22.el8
UNITY680 storage
Our hosts have 4 paths for each LUN.
The paths go through 2 different SPs.
When an SP is restarted or fails, you see that 2 paths are lost, but 2 are still available.
Despite having 2 paths available, OLVM gives a timeout message: "vdsm command spmstatusvds fails" and the server that has the SPM role is marked as non-responsive and is restarted with the fence.
Then the rest of the servers go down 1 by 1 until the whole farm is restarted.
Is there a way to prevent this? It does not make sense as we have redundant paths.