Hi,
I'm having lots of trouble starting my zfs /var partition as part of boot, after an upgrade to Bullseye. I can manually import the partition with
zpool import -a and the status of the pool says "no known data errors".
journalctl | grep zfs gives the following error;
udevadm => systemd-udev-settle.service is deprecated, Please fix zfs-import-cache.service, zfs-load-module.service not to pull it in.
systemd-modules-load[466]: Inserted module 'zfs'
systemd[1]: zfs-import-cache.service: Job zfs-import-cache.service/start failed with result 'dependency'
systemd[1]: zfs-load-module.service: Job zfs-load-module.service/start failed with result 'dependency'
I thought maybe /etc/zfs/zpool_tank.cache had got corrupted, so I set it to none,moved the file, rebooted, reset it to /etc/zfs/zpool_tabk.cache and reboot but no joy.
I've tried adding "ExecPreStart=/bin/sleep 5" to the service section of /lib/systemd/systemzfs-import-cache
Did a systemctl daemon-reload
rebooted
I've checked that zfs is listed in /etc/modules-load.d/modules.conf and the fact that you don't have to modprobe once started,suggests that it's working.
Unfortunately no change. Now I'm wondering did the system try to load the pool after the kernel module, but I don't want to hack around any more and break things. Note that this used to work under Debian 10 and I've not changed the partition.
It seems to be related to this; but I don't want to change the system any more, without solid advice.
If anyone can offer any help, I'd be very grateful,
James