Skip to main content

Alpha This is a new website theme. Help me improve it and give your feedback (opens in a new tab).

OpenBSD Worrying RAID

Published:

Tags:

Sysadmin Openbsd Raid Storage Planet FSFE
This blog post is more than two years old. It is preserved here in the hope that it is useful to someone, but please be aware that links may be broken and that opinions expressed here may not reflect my current views. If this is a technical article, it may no longer reflect current best practice.

I wanted to move a couple of USB hard drives from one OpenBSD machine to another. They are configured with softraid(4) as RAID 1 (mirrored). When I plugged the drives into the new machine though, nothing happened with softraid. This was pretty worrying.

Both the drives showed in dmesg output so the issue was specifically to do with softraid. The man page for bioctl(8) talks about -c creating a “new” RAID device which sounded a little too destructive. I asked for help in #openbsd and apparently the language in the man page is misleading. The -d flag has recently been updated to say “detach” rather than “delete” to try to address this.

I went for it and did:

bioctl -c 1 -l /dev/sd5a,/dev/sd6a

It worked and I got the dmesg output:

softraid0: sd3 was not shutdown properly
sd7 at scsibus4 targ 1 lun 0: <OPENBSD, SR RAID 1, 006>
sd7: 7630885MB, 512 bytes/sector, 15628052512 sectors
softraid0: volume sd7 is roaming, it used to be sd3, updating metadata
softraid0: roaming device sd2a -> sd6a
softraid0: roaming device sd1a -> sd5a

I guess if it’s not cleanly shutdown it doesn’t just automatically set up the RAID device again, it could also have been the renumbering that stopped it.

Mounting the device:

mount_ffs: /dev/sd7i on /external: filesystem must be mounted read-only; you may need to run fsck
WARNING: R/W mount of /mnt denied.  Filesystem is not clean - run fsck

I guess that was expected. An fsck later and everything was working again.

Backup your stuff.

Contact