U2 + SCSI multipack (6) + Solaris 9 - filesystems corrupt

From: rossman@columbia.edu
Date: Fri Nov 19 2004 - 13:39:17 EST


Hi,

I have an Ultra2 with two internal disks and a six-disk multipack
connected to the same internal SCSI channel (multipack is, of
course, set for targets 9-14).

I HAD three pairs of two-disk mirrors with intact filesystems
prior to my having to change things on the root disks (internal).
After I made some rather dramatic changes, I realized I had
blown away the metadatabases for everything. OK, I thought, I
can rebuild everything -- I should not have destroyed any
data out on that multipack.

Yet, I have problems with two of the three pairs of mirrored
disks out on the multipack. One pair is fine (well, after I
remirrored the otherwise intact filesystem on one disk of the
pair), but the other two give me some nasty fsck errors.

I'm wondering if this somehow has to do with journalled filesystems
in S9 vs non-journalled?... here's what I get if I fsck any of
the four remaining individual component slices on which there was
once a file system:

(first, the meta-database as it is now):

jeeves# metastat -p
d50 -m d51 d52 1
d51 1 1 c0t9d0s0
d52 1 1 c0t10d0s0
d3 -m d13 d23 1
d13 1 1 c0t0d0s3
d23 1 1 c0t1d0s3
d1 -m d11 d21 1
d11 1 1 c0t0d0s1
d21 1 1 c0t1d0s1
d0 -m d10 d20 1
d10 1 1 c0t0d0s0
d20 1 1 c0t1d0s0
jeeves#

t9 and t10 are the "good" disks/filesystems in the multipack.
t11 through t14 don't seem to have fsck-able file systems anymore.
I get the following:

jeeves# fsck /dev/dsk/c0t11d0s0
** /dev/dsk/c0t11d0s0

CANNOT READ: BLK 71000096
CONTINUE? y

THE FOLLOWING SECTORS COULD NOT BE READ:
BAD SUPER BLOCK: BAD VALUES IN SUPER BLOCK
USE AN ALTERNATE SUPER-BLOCK TO SUPPLY NEEDED INFORMATION;
eg. fsck [-F ufs] -o b=# [special ...]
where # is the alternate super block. SEE fsck_ufs(1M).
jeeves#
jeeves# fsck -o b=32 /dev/dsk/c0t11d0s0
Alternate super block location: 32.
** /dev/dsk/c0t11d0s0

CANNOT READ: BLK 71000096
CONTINUE? y

THE FOLLOWING SECTORS COULD NOT BE READ:
** Last Mounted on
** Phase 1 - Check Blocks and Sizes

CANNOT READ: BLK 8691520
CONTINUE? y

THE FOLLOWING SECTORS COULD NOT BE READ:

CANNOT READ: BLK 33512864
CONTINUE? y

THE FOLLOWING SECTORS COULD NOT BE READ:

CANNOT READ: BLK 30278912
CONTINUE? y

THE FOLLOWING SECTORS COULD NOT BE READ:

CANNOT READ: BLK 16504288
CONTINUE? ^C
jeeves#

Anyone have any ideas what is actually going on here? The same
symptoms exist for all four of the remaining disks in the multi-
pack. Any ideas on fixes?

Thanks,

Ken Rossman
rossman@columbia.edu
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:29:44 EDT