[SOLVED] zfs volume not listed in storage when creating new vm or lxc after upgrade to proxmox cluster to v8.4.11

[Sundude]

Hi
Existing vm and lxc are running fine (and links to the local-zfs storage). but when trying to create an new vm or lxc using local zfs mirror storage it does not show.
the setup is 3 promox hosts in an single cluster (for easier management, all using local storage only, no ha or ceph) It seems to have happened after upgrade to 8.4.11 (from 8.4.7 as far as i remember). it the same issue on all 3 proxmox hosts. Hope you can give some hints want i need to do investigate. anyway right not i cannot create any new vm or lxc using the local-zfs pool

Datacenter storage local-zfs

local-zfs points to /rpool/data , it seems correct?


new vm window as you can see the local-zfs does not exists and it only shows 2 of the 4 configured datacenter storage items. (note the cluster seems to be ok no errors found and new items like an network vlan sdn creation replicates fine across the 3 cluster hosts.



also preseeing the local-zfs host volumen tree also works fine (and all existing lxc/vm runs fine using the local-zfs)

Example lxc using the local-zfs volume.



Looking to zfs there seem to be some with the cache? not sure this is related to this issue? But here are some cli status commands seems there are something wrong but not sure what to do about it?

-----
systemctl status zfs-import.service zfs-import-cache.service
○ zfs-import.service
Loaded: masked (Reason: Unit zfs-import.service is masked.)
Active: inactive (dead)

○ zfs-import.service
Loaded: masked (Reason: Unit zfs-import.service is masked.)
Active: inactive (dead)

○ zfs-import-cache.service - Import ZFS pools by cache file
Loaded: loaded (/lib/systemd/system/zfs-import-cache.service; enabled; preset: enabled)
Active: inactive (dead)
Condition: start condition failed at Sat 2025-08-23 13:03:59 CEST; 53min ago
Docs: man:zpool(8)

Aug 23 13:03:59 pvec1 systemd[1]: zfs-import-cache.service - Import ZFS pools by cache file was skipped because of an unmet condition check (ConditionFileNotEmpty=/etc/zfs/zpool.cache)

Note:the zpool.cache file does not exists.
----
journalctl -b0 | grep -i zfs -C 2

Aug 23 13:03:57 pvec1 systemd[1]: Starting ifupdown2-pre.service - Helper to synchronize boot up for ifupdown...
Aug 23 13:03:57 pvec1 systemd[1]: Starting systemd-udev-settle.service - Wait for udev To Complete Device Initialization...
Aug 23 13:03:57 pvec1 udevadm[966]: systemd-udev-settle.service is deprecated. Please fix zfs-import-cache.service, zfs-import-scan.service not to pull it in.

Aug 23 13:03:59 pvec1 systemd[1]: zfs-import-cache.service - Import ZFS pools by cache file was skipped because of an unmet condition check (ConditionFileNotEmpty=/etc/zfs/zpool.cache).
Aug 23 13:03:59 pvec1 systemd[1]: Starting zfs-import-scan.service - Import ZFS pools by device scanning...
Aug 23 13:03:59 pvec1 zpool[1040]: no pools available to import
Aug 23 13:03:59 pvec1 systemd[1]: Finished zfs-import-scan.service - Import ZFS pools by device scanning.
Aug 23 13:03:59 pvec1 systemd[1]: Reached target zfs-import.target - ZFS pool import target.
Aug 23 13:03:59 pvec1 systemd[1]: Starting zfs-mount.service - Mount ZFS filesystems...
Aug 23 13:03:59 pvec1 systemd[1]: Starting zfs-volume-wait.service - Wait for ZFS Volume (zvol) links in /dev...
Aug 23 13:03:59 pvec1 zvol_wait[1087]: Testing 16 zvol links
Aug 23 13:03:59 pvec1 zvol_wait[1087]: All zvol links are now present.
Aug 23 13:03:59 pvec1 systemd[1]: Finished zfs-volume-wait.service - Wait for ZFS Volume (zvol) links in /dev.
Aug 23 13:03:59 pvec1 systemd[1]: Reached target zfs-volumes.target - ZFS volumes are ready.
Aug 23 13:03:59 pvec1 systemd[1]: Finished zfs-mount.service - Mount ZFS filesystems.
Aug 23 13:03:59 pvec1 systemd[1]: Reached target local-fs.target - Local File Systems.
Aug 23 13:03:59 pvec1 systemd[1]: Starting apparmor.service - Load AppArmor profiles...
--------

zpool status
pool: rpool
state: ONLINE
scan: scrub repaired 0B in 00:03:06 with 0 errors on Sun Aug 10 00:27:07 2025
config:

NAME STATE READ WRITE CKSUM
rpool ONLINE 0 0 0
mirror-0 ONLINE 0 0 0
nvme-eui.0025382141421be1-part3 ONLINE 0 0 0
nvme-nvme.1e4b-5450424632343039303930313330313031333833-5445414d20544d3846504b30303254-00000001-part3 ONLINE 0 0 0

errors: No known data errors




Best Regards
Jan

 

[Sundude]

Seem to solved now, fixed a lot of small stupid mistakes and the zpool issues after the backup.
1. dns was referring to an dns server location in a vm/docker in one of the promox hosts.. Adding static hosts entries for the 3 proxmox hosts was added to avoid similar catch 22
2. I overlook the boot image part not reading what it said (because i was so focused on the zfs error) i had only added the images/iso the nfs and local and not to the local-zfs
3. zfs cache issue was that there was no file on the host (not sure why).
Used the zpool list (or zpool status) to get the correct pool name, my zfs pool name was rpool
executed the
zpool set cachefile=/etc/zfs/zpool.cache rpool
now the issue was gone in systemctl status zfs-import-cache.service

Well i always seem to learn something new moving from windows and esxi over the last 1 year,
(hmm not sure how to mark this as solved).

/jan