Solved: Cant Migrate a vm on shared storage in a newly created cluster.

[nktech1135]

Hello all.
I've just set up a cluster and am trying to move vms around. The vm's disk is on my shared storage pool nfs1.
If anyone has any ideas on what i did wrong, please let me know.
Here are the details.

root@sm01:~# qm config 107
agent: 1
boot: c
bootdisk: scsi0
cipassword: **********
cores: 2
ipconfig0: ip=dhcp
memory: 2048
name: dvr
net0: virtio=06:7F:39C:60:28,bridge=vmbr0
numa: 0
onboot: 1
ostype: l26
scsi0: nfs1:107/vm-107-disk-0.qcow2,size=32972M
scsihw: virtio-scsi-pci
serial0: socket
smbios1: uuid=ac52b189-4718-4ea9-a5ec-98965476c7c8
sockets: 1
sshkeys: ssh-rsa%20AAAAB3NzaC1yc2EAAAADAQABAAABAQCnqFVxZn135ah6RJIKIopxoXFHmtWNufQml7lO%2BE4sxpNodaPJ2IZxnEzs%2FBJ5fdRWnI3eUASGiEJrPTzuDuySW7td2yN3Ft3YcPHZk8N%2Fj55pXSjO5%2BQJ1qU698K5x9f8d8zZkxGE6HrxdZkUFNLalW4kTYwupE2Fo2sseFWMPR5bTInkuKEd%2BV4Q1szuWgXMwPE5ibobuVlajmiiuHrYP%2B2k%2F6%2BQdmZoqUA4j61egxizjMwIManfCJ%2F8Mlq0rZO1Ft2EQuOfL3W2hPiy8YahTK1zXjcAs5RoAALZtj8wraNeAa0VaGxQk%2FnCAa1KNHHksWj3Y7cw6nKY3TsA%2BNGf%20imported-openssh-key
vga: serial0
vmgenid: a493aeda-90ba-448f-9e50-430b72d1ee8b
root@sm01:~# qm migrate 107 cdpve --online
2020-11-11 08:23:38 starting migration of VM 107 to node 'cdpve' (10.0.0.1)
zfs error: cannot open 'dpool': no such pool

zfs error: cannot open 'dpool': no such pool

2020-11-11 08:23:38 ERROR: Failed to sync data - could not activate storage 'dpool', zfs error: cannot open 'dpool': no such pool
2020-11-11 08:23:38 aborting phase 1 - cleanup resources
2020-11-11 08:23:38 ERROR: migration aborted (duration 00:00:00): Failed to sync data - could not activate storage 'dpool', zfs error: cannot open 'dpool': no such pool
migration aborted
root@sm01:~# pvesm status
zfs error: cannot open 'dpool': no such pool

zfs error: cannot open 'dpool': no such pool

could not activate storage 'dpool', zfs error: cannot open 'dpool': no such pool

zfs error: cannot open 'mail': no such pool

zfs error: cannot open 'mail': no such pool

could not activate storage 'mail', zfs error: cannot open 'mail': no such pool

Name Type Status Total Used Available %
dpool zfspool inactive 0 0 0 0.00%
dpool1 zfspool active 9442950520 432563752 9010386768 4.58%
local dir active 183902848 5723520 178179328 3.11%
local-zfs zfspool active 233931328 55751984 178179344 23.83%
mail zfspool inactive 0 0 0 0.00%
nfs1 nfs active 183903232 5723136 178180096 3.11%
root@sm01:~# pvecm status
Cluster information
-------------------
Name: bks1
Config Version: 3
Transport: knet
Secure auth: on

Quorum information
------------------
Date: Wed Nov 11 09:54:00 2020
Quorum provider: corosync_votequorum
Nodes: 2
Node ID: 0x00000002
Ring ID: 1.36
Quorate: Yes

Votequorum information
----------------------
Expected votes: 2
Highest expected: 2
Total votes: 2
Quorum: 2
Flags: Quorate

Membership information
----------------------
Nodeid Votes Name
0x00000001 1 10.0.0.1
0x00000002 1 10.0.0.2 (local)
root@sm01:~# cat /etc/pve/storage.cfg
dir: local
path /var/lib/vz
content vztmpl,iso,backup

zfspool: local-zfs
pool rpool/data
content images,rootdir
sparse 1

zfspool: dpool
pool dpool
content rootdir,images
sparse 0

zfspool: mail
pool mail
content rootdir,images
mountpoint /mail
sparse 1

nfs: nfs1
export /mnt/nfs1
path /mnt/pve/nfs1
server 10.2.2.21
content images,rootdir
options vers=4.2

zfspool: dpool1
pool dpool1
content images,rootdir
mountpoint /dpool1
sparse 0

 

[mira]

The storage 'dpool' is not available on your target node. If a storage is only available on some nodes of the cluster, you can limit it to those in the options.

 

[nktech1135]

I get the fact that dpool isn't available on the target node. However, since the vm config doesn't reference dpool i'm confused as to why it matters?
Am i missing something in the config that's still referencing dpool?

Thanks.

 

[mira]

When migrating it activates the storages, which can't be done on the target node in this case, because the storage 'dpool' is referenced in the storage config, but is not available on the node.

 

[nktech1135]

Forgive me for missing something here.


you say "because the storage 'dpool' is referenced in the storage config, but is not available on the node."

Why does this matter since the vm itself doesn't reference dpool in any way.


You also say "If a storage is only available on some nodes of the cluster, you can limit it to those in the options."

Can you elaborate on this or send me a guide on setting up the mentioned limitations?

Thanks.

 

[narrateourale]

If you edit a storage, you can limit it to certain nodes only. In the edit window top right. This way the PVE tooling will know on which node it can expect which storage and the migration should work.

 

[nktech1135]

I'm not sure why this didn't click until now. That works great and fixed my migration works as well now.