Hi Guys, I installed the latstest updates last friday on my cluster to:
Linux 7.0.14-15-pve (2026-08-26T14:21Z)
But ran into some strange issues after that:
Rebooting a node migrates the HA VMs to the other node, no issue, but when the node is online again migrating back fails with an error that states that there is no route to host,
task started by HA resource agent
ssh: connect to host xxx.xxx.xxx.xxx port 22: No route to host
TASK ERROR: command '/usr/bin/ssh -e none -o 'BatchMode=yes' -o 'HostKeyAlias=yyyy' -o 'UserKnownHostsFile=/etc/pve/nodes/yyyy/ssh_known_hosts' -o 'GlobalKnownHostsFile=none' root@xxx.xxx.xxx.xxx pvecm mtunnel -migration_network xxx.xxx.xxx.xxx/24 -get_migration_ip' failed: exit code 255
After soms troubleshooting I added a delay in the pve-ha-lrm of 15 seconds, that fixed it, For reference, a friend of mine has the same setup as me (2 nodes and a Qdevice) but running on different hardware, and he has no issues.
Other problem since the upgrade is that 1 host keeps disappearing from the cluster GUI, everything is marked with "?" , but all VMs run etc. When restarting pvestatd it came back in de GUI. Investigating the logging of that service I found that I have errors stating that my PBS server storage can't be reached pvestatd[1599]: ProxmoxBackupServer: error fetching datastores - 500 Can't connect to pbs.xxxxxx:8007 (Temporary failure in name resolution), also sometime other NFS shares.
That wasn't the case before the upgrade.
After some searching I found this old thread: https://forum.proxmox.com/threads/proxmox-freezes-random-no-clue-in-the-log.89212/#:~:text=* log.zip. 406.3 KB · Views: 4.
I have these NICs:
Intel Corporation Ethernet Controller I226-V (rev 04)
6 per node
EDIT: I reverted to the kernel I was running before (and removed the delay), but issues are still there.
Linux 7.0.14-15-pve (2026-08-26T14:21Z)
But ran into some strange issues after that:
Rebooting a node migrates the HA VMs to the other node, no issue, but when the node is online again migrating back fails with an error that states that there is no route to host,
task started by HA resource agent
ssh: connect to host xxx.xxx.xxx.xxx port 22: No route to host
TASK ERROR: command '/usr/bin/ssh -e none -o 'BatchMode=yes' -o 'HostKeyAlias=yyyy' -o 'UserKnownHostsFile=/etc/pve/nodes/yyyy/ssh_known_hosts' -o 'GlobalKnownHostsFile=none' root@xxx.xxx.xxx.xxx pvecm mtunnel -migration_network xxx.xxx.xxx.xxx/24 -get_migration_ip' failed: exit code 255
After soms troubleshooting I added a delay in the pve-ha-lrm of 15 seconds, that fixed it, For reference, a friend of mine has the same setup as me (2 nodes and a Qdevice) but running on different hardware, and he has no issues.
Other problem since the upgrade is that 1 host keeps disappearing from the cluster GUI, everything is marked with "?" , but all VMs run etc. When restarting pvestatd it came back in de GUI. Investigating the logging of that service I found that I have errors stating that my PBS server storage can't be reached pvestatd[1599]: ProxmoxBackupServer: error fetching datastores - 500 Can't connect to pbs.xxxxxx:8007 (Temporary failure in name resolution), also sometime other NFS shares.
That wasn't the case before the upgrade.
After some searching I found this old thread: https://forum.proxmox.com/threads/proxmox-freezes-random-no-clue-in-the-log.89212/#:~:text=* log.zip. 406.3 KB · Views: 4.
I have these NICs:
Intel Corporation Ethernet Controller I226-V (rev 04)
6 per node
EDIT: I reverted to the kernel I was running before (and removed the delay), but issues are still there.
Last edited: