Sorry for the delay in this summary;
>In list.sun-managers I wrote:
>
>>We have been getting lots of disk error messages of the form
>>xy1d: read restore (drive fault) -- blk #478624, abs blk #478624
>>xy1d: read restore (drive fault) -- blk #346304, abs blk #346304
>
>>with various blocks in the last few days. The last was followed by
>>xy1d: read retry (header not found) -- blk #476912, abs blk #476912
>>xy1d: read failed (drive not ready) -- blk #476912, abs blk #476912
>>vmunix: xy1: offline
>>vmunix: panic: iinactive
>>vmunix: syncing file systems... panic: iinactive
>>and a reboot. Does this mean the disk is dying, or is there
>>another cause?
There were many helpful replies.
Ken Rossman <ken@shibuya.cc.columbia.edu>
sent me a script to pinpoint the bad blocks which were all over the
disk. We replaced the controller, and the messages stopped. All was
well for about a week, then the system suddenly crashed, and refused
to reboot. After booting diskless, xy0a would not fsck cleanly, and
newfs didn't help. Restore bombed after creating directories, before
restoring files. fsck gives lots of errors on xy1d, but the files ARE
ok (or seem to be) Errors of the form PARTIALLY ALLOCATED INODE, or
INVALID FILE TYPE 43, etc. The miniroot can be copied, apparently to
xy0b but will not boot. Starts to, hits REG 80 invalid. System runs
fine diskless, and the files on xy1d appear to be ok, even though fsck
complains loudly.
I am totally confused. (SUN 3/160 running 4.1 )
One respondent (whose note I lost SORRY) had the same, problem , also
replaced controller, success, and a week later the disk overwrote its
label. I didn't findout what happend. If you are listening Tell me.
Thanks to all
Vince <vince@math.uconn.edu>
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:16 CDT