UPON LOST QUORUM (not indicated in the GUI), the attempt to access the node's OWN console ends up in generic:
This is clearly desired behaviour for VMs running on a node that lost quorum, however GUI provides no indication that there is a lost quorum situation, the natural progression of troubleshooting is then to get the node's root console (of which the GUI is being accessed), this however results in the above error just as well.
The only temporary way to get a root shell is to log in NOT through the GUI (tty or ssh), this is even if one needs to continue with the documented "pvecm expected $count" to artificially get the node to accept manual intervention.
Posting it here so that others can find it when troubleshooting, will file Bugzilla reports later on. I believe it would be highly beneficial to show:
1) Lost quorum in GUI
2) Allow console access to the same node of which GUI is accessed in such situation
Code:
failed waiting for client: timed out
TASK ERROR: command '/usr/bin/termproxy 5900 --path /nodes/pve7 --perm Sys.Console -- /bin/login -f root' failed: exit code 1This is clearly desired behaviour for VMs running on a node that lost quorum, however GUI provides no indication that there is a lost quorum situation, the natural progression of troubleshooting is then to get the node's root console (of which the GUI is being accessed), this however results in the above error just as well.
The only temporary way to get a root shell is to log in NOT through the GUI (tty or ssh), this is even if one needs to continue with the documented "pvecm expected $count" to artificially get the node to accept manual intervention.
Posting it here so that others can find it when troubleshooting, will file Bugzilla reports later on. I believe it would be highly beneficial to show:
1) Lost quorum in GUI
2) Allow console access to the same node of which GUI is accessed in such situation