established zpool inaccessible after reboot...

[erll]

i currently have zfs-macOS-2.2.3rc5 installed on a Mac Mini M1 running macOS Sequoia 15.2. prior to a reboot today, i had a functioning zpool up and running (that auto-mounted after previous reboots). today, i installed the latest version of Paragon Software's extFS for Mac to test. after rebooting to complete the installation, the zpool did not auto-mount. i then attempted to import the zpool via Terminal, but got the message "no pools available to import." all of the disks associated with the zpool show up in diskutil.

does anyone have an idea of what the issue may be?

[erll]

after investigating a few other threads, i was able to make some progress by using the "zpool import -d /dev" command. the zpool is listed in the output, but with the error message "the pool cannot be imported using its name or numeric identifier."

[erll]

i was able to import the zpool using the "zpool import -d /dev -a" command. can anyone recommend how to avoid having to do this again in the future? or, for that matter, tell me exactly what went wrong in the first place?

[Sharko]

I don't have any specific knowledge about what Paragon Software's ExtFS does, but the symptoms you describe suggest to me that ExtFS looked at your disks, and somehow altered your ZFS installation. It might have altered your gpt table on the disk that had a zpool on it; there are standardized codes for each type of partition (HFS+, ext4, zfs, etc), and perhaps ExtFS saw that there was a disk with a non-Apple-native partition on it (your zpool disk), so it just said "Well, I'm going to claim it as ext4!"

That would be an unfriendly action to take without warning you, but it fits the facts as you report them.

MacOS has a gpt utility that you can use to inspect the gpt table (in this example querying /dev/disk3):

Code: Select all

sudo gpt -r show disk3


You can find a listing of what the different entry constants correspond to at this link:

https://wiki-gateway.eudic.net/wikipedi ... Table.html