[SOLVED] upgrade to 6.1-8 has broken zfs mount

[hinchles]

I have a zfs pool created called VMStorage and its been mapped (for last 2 years) to /VMStorage.
I've just updated to latest version and rebooted but now the system is no longer mounting the pool.
The system knows the pool is there, the web UI shows the pool and allows me to access its content. However doing ls /VMStorage/ returns nothing and now I can't start any of my VM's up but yet the pool thinks its valid.

I have no idea how to fix this at the moment as everything I've found has suggested that things like zfs list should be blank etc when in this status.

root@kessel:/# cat /etc/pve/storage.cfg
dir: local
        path /var/lib/vz
        content backup,vztmpl,iso

lvmthin: local-lvm
        thinpool data
        vgname pve
        content images,rootdir

zfspool: VMStorage
        pool VMStorage
        content rootdir,images
        nodes kessel

root@kessel:/etc/pve# zfs list
NAME                          USED  AVAIL     REFER  MOUNTPOINT
VMStorage                    45.9G   814G      112K  /VMStorage
VMStorage/subvol-100-disk-0  38.7G  26.4G     23.6G  /VMStorage/subvol-100-disk-0
VMStorage/subvol-101-disk-0  1.47G  6.53G     1.47G  /VMStorage/subvol-101-disk-0
VMStorage/subvol-102-disk-0  2.28G  78.1G     1.94G  /VMStorage/subvol-102-disk-0
VMStorage/subvol-103-disk-0  1.79G  18.8G     1.23G  /VMStorage/subvol-103-disk-0
VMStorage/subvol-104-disk-0   481M  9.53G      481M  /VMStorage/subvol-104-disk-0
VMStorage/subvol-104-disk-1   656M  99.4G      656M  /VMStorage/subvol-104-disk-1
VMStorage/subvol-105-disk-0   498M  7.51G      498M  /VMStorage/subvol-105-disk-0

root@kessel:/etc/pve# pvesm status
Name             Type     Status           Total            Used       Available        %
VMStorage     zfspool     active       902036868        48131068       853905800    5.34%
local             dir     active        58429568        35751420        22678148   61.19%
local-lvm     lvmthin     active       147275776               0       147275776    0.00%

root@kessel:/tmp# df
Filesystem           1K-blocks     Used Available Use% Mounted on
udev                  16398880        0  16398880   0% /dev
tmpfs                  3284712     9044   3275668   1% /run
/dev/mapper/pve-root  58429568 35752456  22677112  62% /
tmpfs                 16423556    43680  16379876   1% /dev/shm
tmpfs                     5120        0      5120   0% /run/lock
tmpfs                 16423556        0  16423556   0% /sys/fs/cgroup
/dev/sde2               523248      312    522936   1% /boot/efi
/dev/fuse                30720       20     30700   1% /etc/pve
tmpfs                  3284708        0   3284708   0% /run/user/0

 

[aaron]

What does the command mount give you?

Does a directory /VMStorage exist? ls -la /VMStorage

 

[hinchles]

I actually fixed this as part of another thread about containers not running after the update, a forum member in there suggested unmount manually export the pool then hard delete the mount point on the server and import the pool.
This fixed it properly.

To clarify your questions above though as it still may be helpful for you.

So in the unmounted state the folder /vmstorage existed but was empty, doing zfs mount VMStorage then meant doing ls inside /vmstorage showed me the contents of the pool ie: all the subvol folders but they only contained a dev folder and that was empty.
This was why the containers also wouldn't run because the mount points were empty.

Some how exporting and re-importing fixed the issue and everything remounted just fine. doing df/ls etc all exactly the same as before but this time going into the subvol folder shows the entire os file structure for the container not just an empty dev folder.

 

[aaron]

Good to hear. Can you please mark the thread as solved then? (edit the first post and select the prefix)