Hi,
I have a PVE7 cluster with 3 nodes, and SAN storage by fiber channel for storing VMs.
With the old kernel (5.15.53-1-pve), all is perfect : nodes can access to LUNs and VMs migration between nodes is OK.
I updated the first node 2 weeks ago, to kernel 5-15-64-1-pve, and SAN is not working from this node, VMs migration is impossible. I've tried to update yesterday with the new kernel (5-15-64-1-pve & 5.15.74-1-pve), same problem. But if I force boot with 5.15.53-1-pve, all is working - so I suppose it's a problem with kernel and not a specific package recently updated.
When I start VM migration :
When I check my SAN storage from the node, no disk is displayed on "VM disks" (and the icon comes with the question mark...).
No changes on multipath.conf and wwids files, it's the sames than other nodes.
On SAN :
- Status is degraded
- LUNs and network address : OK
- No initiator paths
- Initiators : "The initiator does not have any logged in initiator paths. Check the connection between the initiator and the storage system"
So, what's wrong with new kernels? How can I retreive initiator paths other than reboot the node with an old kernel?
Thanks !
I have a PVE7 cluster with 3 nodes, and SAN storage by fiber channel for storing VMs.
With the old kernel (5.15.53-1-pve), all is perfect : nodes can access to LUNs and VMs migration between nodes is OK.
I updated the first node 2 weeks ago, to kernel 5-15-64-1-pve, and SAN is not working from this node, VMs migration is impossible. I've tried to update yesterday with the new kernel (5-15-64-1-pve & 5.15.74-1-pve), same problem. But if I force boot with 5.15.53-1-pve, all is working - so I suppose it's a problem with kernel and not a specific package recently updated.
When I start VM migration :
2022-11-22 16:39:53 starting migration of VM 116 to node 'pve-prod-01' (172.16.240.101)
2022-11-22 16:39:53 starting VM 116 on remote node 'pve-prod-01'
2022-11-22 16:39:55 [pve-prod-01] can't activate LV '/dev/vg_lun_01/vm-116-disk-0': Cannot process volume group vg_lun_01
2022-11-22 16:39:55 ERROR: online migrate failure - remote command failed with exit code 255
2022-11-22 16:39:55 aborting phase 2 - cleanup resources
2022-11-22 16:39:55 migrate_cancel
2022-11-22 16:39:56 ERROR: migration finished with problems (duration 00:00:05)
TASK ERROR: migration problems
Volume group "vg_lun_01" not found
TASK ERROR: can't activate LV '/dev/vg_lun_01/vm-116-disk-0': Cannot process volume group vg_lun_01
When I check my SAN storage from the node, no disk is displayed on "VM disks" (and the icon comes with the question mark...).
No changes on multipath.conf and wwids files, it's the sames than other nodes.
On SAN :
- Status is degraded
- LUNs and network address : OK
- No initiator paths
- Initiators : "The initiator does not have any logged in initiator paths. Check the connection between the initiator and the storage system"
So, what's wrong with new kernels? How can I retreive initiator paths other than reboot the node with an old kernel?
Thanks !
