Two node cluster, migration fails with local pve-zsync drive

[iMx]

Hi there,

Apologies if this has already been answered somewhere, I was not able to find a conclusive answer.

Long story short - I have a 2 node cluster, using containers only, each node has SSD based zpool for the main disk storage and then an HDD zpool which I was planning to use for pve-zsync.

I've configured pve-zsync to take hourly/weekly snapshots to the HDD zpool, however this then seems to cause the migration to fail. Is there something I should set so it ignores the HDD zpool, if this is possible? The containers disks are stored only on the SSD zpool.

dir: local
        path /var/lib/vz
        content backup,vztmpl,iso

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

zfspool: backup-pm001
        pool backup-pm001
        content images,rootdir
        nodes pm001

zfspool: backup-pm004
        pool backup-pm004
        content images,rootdir
        nodes pm004

Error when trying to migrate:

()
2019-02-14 14:23:28 shutdown CT 104
2019-02-14 14:23:30 starting migration of CT 104 to node 'pm001' (10.99.0.10)
2019-02-14 14:23:30 ERROR: storage 'backup-pm004' is not available on node 'pm001'
2019-02-14 14:23:30 aborting phase 1 - cleanup resources
2019-02-14 14:23:30 start final cleanup
2019-02-14 14:23:30 start container on source node
2019-02-14 14:23:31 ERROR: migration aborted (duration 00:00:03): storage 'backup-pm004' is not available on node 'pm001'
TASK ERROR: migration aborted

I tried to disable the pve-zsync job, but this doesn't seem to help - is it possible to use local storage with pve-zsync and retain migration?

Many thanks,

 

[iMx]

Should have RTFM-ed... set to disable, seems to then work. Apologies.

dir: local
        path /var/lib/vz
        content backup,vztmpl,iso

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

zfspool: backup-pm001
        pool backup-pm001
        content images,rootdir
        nodes pm001
        disable

zfspool: backup-pm004
        pool backup-pm004
        content images,rootdir
        nodes pm004
        disable

 

[tschanness]

ERROR: migration aborted (duration 00:00:03): storage 'backup-pm004' is not available on node 'pm001'

-> you need to specify the correct datastore

 

[iMx]

ERROR: migration aborted (duration 00:00:03): storage 'backup-pm004' is not available on node 'pm001'

-> you need to specify the correct datastore

You select a node when performing Migrate of a container through the UI, there is no option to specify the datastore - it syncs to the replicated datastore.

Setting my dedicated pve-zsync volumes to disabled removes them from the UI, allows the migration to complete but still seems to allow pve-zsync to run.