[SOLVED] VM migration in pools CEPH

P

piviul

Active Member
Mar 19, 2020
50
3
28
I have just configured CEPH and I'm trying to test it. Migrating the HD of the 103 VM from ZFS pool to the CEPH pool has been completed successfully. Even migrating the 103 VM from one node to another has been completed successfully so all seems to work correctly... but now when I try to migrate back to the original node or another node I get a task error:
2021-03-06 05:44:36 starting migration of VM 103 to node 'pve02' (192.168.70.32)
2021-03-06 05:44:36 ERROR: Failed to sync data - storage 'zfspool' is not available on node 'pve02'
2021-03-06 05:44:36 aborting phase 1 - cleanup resources
2021-03-06 05:44:36 ERROR: migration aborted (duration 00:00:00): Failed to sync data - storage 'zfspool' is not available on node 'pve02'
TASK ERROR: migration aborted
Click to expand...
The VM 103 has nothing to do now with ZFS: why the migration has to sync zfs storage? Furthermore in pve02 node zfspool is available! I'm very confused...

Piviul
 
piviul said:
The VM 103 has nothing to do now with ZFS: why the migration has to sync zfs storage? Furthermore in pve02 node zfspool is available! I'm very confused...
Click to expand...
Are you sure it hasn't, qm config <vmid>? You need to specify on the storage tab (DC -> Storage -> zfspool) that all nodes with that zfs pool do also have a storage.
 
# qm config 103
boot: dcn
bootdisk: scsi0
cores: 4
ide2: none,media=cdrom
memory: 32768
name: intranet-new
net0: virtio=9E:09:2E:2D:6F:DA,bridge=vmbr0,firewall=1
numa: 0
onboot: 1
ostype: l26
scsi0: ceph-vm:vm-103-disk-0,cache=writeback,size=150G
scsihw: virtio-scsi-pci
smbios1: uuid=906b3615-a3bf-4ee4-84b2-304110356850
sockets: 4
vmgenid: cd8e57c3-ceb2-4b95-89bb-4df31a139884
Click to expand...

Doesn't seems to me to be tied to ZFS anymore... but in effect I've found something wrong in the ZFS configuration of the node that host the VM 103 and the cause it's surely my "newbeness"... if someone can help me to find the problems...

I send you more infos hoping that can help you to understand my problem... On the pve03 node I wouldn't use ZFS so I have created a pool only to remove the red cross on the node ZFS-store... any way on the pve03 node seems that's allright:
root@pve03:~# zpool status -v
pool: zfspool
state: ONLINE
config:

NAME STATE READ WRITE CKSUM
zfspool ONLINE 0 0 0
scsi-350000c0f02be8b6c ONLINE 0 0 0

errors: No known data errors
Click to expand...

But if I press the Detail button after selecting zfspool on the Disks->ZFS tab of the pve03 node I get the error
Result verification failed (400)
scan: property is missing and it is not optional
Click to expand...
Furthermore in effect if I go to the pve03 node storage (ZFS-store) I can found 2 items: vm-103-disk0 and vm-103-state-intranet-like; the first one (vm-103-disk-0) seems to be an image of the vm and second one vm-103-state-intranet-like IIRC should be an old snapshot that I have already deleted weeks ago but that seems to be migrated erroneously... If I try to remove them I get the error
Cannot remove image, a guest with VMID '103' exists!
You can delete the image from the guest's hardware pane
Click to expand...
But there is no reference to ZFS on the VMID 103...

I'm feeling confused :confused:

Piviul

Thanks a lot

Piviul
 
piviul said:
I send you more infos hoping that can help you to understand my problem... On the pve03 node I wouldn't use ZFS so I have created a pool only to remove the red cross on the node ZFS-store... any way on the pve03 node seems that's allright:
Click to expand...
Edit your pool in DC -> Storage -> zfspool. You can limit the nodes a storage is configured on.
https://pve.proxmox.com/pve-docs/chapter-pvesm.html#_common_storage_properties

piviul said:
Furthermore in effect if I go to the pve03 node storage (ZFS-store) I can found 2 items: vm-103-disk0 and vm-103-state-intranet-like; the first one (vm-103-disk-0) seems to be an image of the vm and second one vm-103-state-intranet-like IIRC should be an old snapshot that I have already deleted weeks ago but that seems to be migrated erroneously... If I try to remove them I get the error
Click to expand...
You will need to cleanup manually, after the storage situation is fixed.
 
You must log in or register to reply here.
Top Bottom