[SOLVED] HA failed on some VM after node rebooting

[nsc]

Hello,

This morning i tried to update a node (i have a cluster with 5 nodes, identical hardware).

So i update with GUI then reboot, and HA starts to move VM. Icon on VM change with "arrow" from migrate action.

But its failed for 3 of them, all Windows :
- for two of them, HA migrate to another node but i can see STOP / START in a loop (with Windows crashed...) on the node
- for the last one, it was crash in "recovery Windows Menu"

i add to go in HA menu to "disabled" each VM, that stop then restart loop, then VM came back to the node automaticly.

Did i miss something ? Is it better to manually migrate VM than just reboot ?

Best regards

Nsc

Exemple :

 

[nsc]

I've created a script to put the server into maintenance before update / rebooting. This script turns off the local disk VMs and migrates those in shared storage.
Then I have a script that resets to the original state.

This allows a much smoother migration, the script works one VM at a time.

I've used it on my 6 servers and so far so good.