Hello everyone, how are you?


I’d like to share my case: I work in data recovery, and last week we finished cloning a striped ZFS volume consisting of 12 disks, each 4 TB. In the past, I’ve been able to recover volumes using only FreeBSD by running zpool import -f, but this time it’s not working.

When I try the following command:

zpool import -F -f -d /dev/md -o readonly=on -o altroot=/mnt/zfs zpool1

it reports that one or more devices are unavailable.

Also, when I check dmesg, the only relevant line at the end is:

pid (1867) (zpool) is attempting to use unsafe AIO request - not logging anymore


I also want to clarify that I’m not working with the client’s original drives, but with clones. I currently don’t have enough spare 4 TB disks available, so the cloned images are stored as files on three 20 TB disks formatted in NTFS. I emulate these images as block devices using:


mdconfig -a -t vnode -f /mnt/taskimage.bin -u 0


I mount the 20 TB disks via ntfs-3g. I’m wondering if this setup may be causing any penalties or issues with ZFS access.

I have also uploaded the zdb logs for each disk, in case they help with diagnosing the issue.

Thanks in advance for any guidance!

I don't see any uploads.

_________________
A backup a day keeps DR away.

I already solved it; it's a proprietary QNAP ZFS, that's why I can't get it to work on FreeBSD.