Sorry for the multipost, this is also on comp.unix.bsd.freebsd, but please..
PLEASE help me with this if you can, I'm in a total state of panic right
now. The short version:
I did a create -f /etc/vinum.conf on an edited configuration file that
already had existing objects in it to activate a new volume, and now I can't
access my previously existing concatenated volume anymore. Major panic..
since that volume was my backup.
The long version:
I created a concatenated volume called backup to temporarily hold a lot of
important files while I convert an existing hardware RAID volume to vinum
(hardware RAID=really cheap FastTrak 100 Lite, which doesn't support custom
stripe sizes). This new backup volume consists of 2 disks of 120GB each (ad2
I created a backup copy of everything on /dev/ar0 onto my new backup volume.
I went about converting the two disks in /dev/ar0 (ad4 and ad6) to a vinum
array by first changing the controller BIOS back to a regular 2-channel
ATA100 with a jumper. Then I destroyed the RAID configuration using
atacontrol. Everything still going according to plan, but now I'm getting to
I edited /etc/vinum.conf to add a striped volume named 'server' to my
system. The syntax of this was correct as the empty striped volume is very
nicely usable. HOWEVER, I typed create -f /etc/vinum.conf to create my new
'server' volume.. which turned out to be a very BAD idea. :o(
The result: a perfectly accessible /dev/vinum/server and a corrupt (?)
I can mount /dev/vinum/backup onto /backup and see its root directory. This
contains a subdirectory named 030625 in which _ALL_ my files are supposed to
be. But whenever I try to cd to it, it fails with an i/o error. An fsck on
/dev/vinum/backup results in partition table errors: "CANNOT FIGURE OUT FILE
SYSTEM PARTITION" just after it reports problems reading a couple of
consecutive blocks near the end of the volume. Please tell me that this can
be fixed... recovering from this could take months otherwise.