Why do New ZFS Pools Created from the GUI Create a SystemD Service that Fails to Import the Pool, even though the Pool is Working? Can this be Fixed?

[SInisterPisces]

This is apparently normal behavior and it's happened since I've been using PVE back in the 7.0.x days, but why does it happen?

What does it mean that systemd fails to "import" a pool that's clearly mounted and usable in PVE?

Why does the rpool not generate this error, but any pool I create does?

 

[fiona]

Hi,
please see https://forum.proxmox.com/threads/zfs-failed-mirror-but-works-fine.143900/post-647175
But sure, the error can be confusing, feel free to open an issue on the bug tracker: https://bugzilla.proxmox.com

 

[fabian]

https://bugzilla.proxmox.com/show_bug.cgi?id=4835

 

[SInisterPisces]

https://bugzilla.proxmox.com/show_bug.cgi?id=4835

Thanks! I posted some of my debug output in the thread you linked here on the forum, and I'm about to ask that the bugzilla bug be re-opened as well, as the current state of things, IMHO, makes it too confusing to even tell how a pool gets imported, and every time a pool is created and destroyed, the failing import service remains in place, so they just pile up.

That can't be working as intended.