I have a fileserver running SuSE 8.0 with 2 IDE Drives and a
SCSI RAID (RAID5). The system boots up to one of the IDE drives,
and the SCSI RAID drive runs as a mounted filesystem (/dev/md0).
(all filesystems are REISERFS).
Problem:
The SCSI RAID drive has crashed (won't boot up), and I need to
recover it (repair it, through some utility).
Further:
The software does not seem to want to mount the disks, even though
the SCSI BIOS (Adaptec) recognizes the disks, and they are listed as
OPTIMAL.
I tried to repair the disk(s) (with reiserfsck), but it produced an
error message:
reiserfs_open: bread failed reading block 2
reiserfs_open: bread failed reading block 16
reiserfs_open: neither new nor old reiserfs form found on /dev/md0
The literature suggested that this error message is hardware related
(either a disk is physically damaged, or maybe the RAID isn't
recognized), but again, the disks are recognized (and therefore,
so is the SCSI Card).
I tried to boot up with SuSE CD 1 in Recovery mode, but not sure
what to do there (i.e., after the root prompt - just run reiserfsck?).
Literature suggests there are plenty of tools to handle disk recovery,
and especially RAID support, I can't find reference to them
anywhere.
The installation of the RAID was trivial (using the YAST2
partitioning GUI window on startup), and yet recovery seems to be a
mystery. This can't be right.
Does anyone know where there are "decent" and "usable" recovery tools?
OR ... does anyone know what "specifically" to do in this case? Can
I re-partition without formatting, and recover the existing data?
Any help is appreciated - I'm sure I'm missing something "obvious", but
can't find "obviously" what to do in the literature, docs, etc.
Thank you.
Robert Amodeo
School of Engineering and Applied Science
University of California, Los Angeles