From: Stephen Ficklin (sficklin@genome.clemson.edu)
Date: Tue Mar 02 2004 - 10:17:50 EST
We've recently been having problems with our StoreEdge L9 tape autoloader
or Solstice 5.5.5, I can't tell which is causing problems. The StoreEdge
L9 is attached to an E450. Recently when it's time for backups to run the
Solstice software indicates that it can't read a label on a tape.
However, the software just recently labeled the tape and usually nothing
has ever been written to the tape. If it can't read the label, the tape
disappears from the list of tapes that can be mounted. Re-doing an
inventory using solstice makes the tape visible for mounting. Sometimes
solstice can read the label on a tape and sometimes it can't. There
doesn't appear to be any obvious reason why it is having problems reading
the labels on the tape. The following log shows an example that happened
today where solstice needed to mount a tape to finish the currently
executing backup. There was only one available tape in the autoloader and
I had to perform an inventory three times before solstice was able to read
the label.
Tue 09:30:20 media info: suggest mounting 000068 on beaker for writing to
pool 'Full'
Tue 09:30:20 media info: loading volume 000068 into /dev/rmt/1hbn
Tue 09:30:46 /dev/rmt/1hbn verify label operation in progress
Tue 09:32:56 media warning: /dev/rmt/1hbn opening: I/O error
Tue 09:32:56 /dev/rmt/1hbn read open error, I/O error
Tue 09:32:57 /dev/rmt/1hbn eject operation in progress
Tue 09:33:35 /dev/rmt/1hbn ejected
Tue 09:34:10 media info: loading volume 000068 into /dev/rmt/1hbn
Tue 09:34:36 /dev/rmt/1hbn verify label operation in progress
Tue 09:36:46 media warning: /dev/rmt/1hbn opening: I/O error
Tue 09:36:46 /dev/rmt/1hbn read open error, I/O error
Tue 09:36:47 /dev/rmt/1hbn eject operation in progress
Tue 09:37:25 /dev/rmt/1hbn ejected
Tue 09:38:00 media info: suggest mounting DR5336 on beaker for writing to
pool 'Full'
Tue 09:39:00 media info: suggest mounting 000068 on beaker for writing to
pool 'Full'
Tue 09:39:01 media info: loading volume 000068 into /dev/rmt/1hbn
Tue 09:39:27 /dev/rmt/1hbn verify label operation in progress
Tue 09:41:32 /dev/rmt/1hbn verified label of 000068
Tue 09:41:34 /dev/rmt/1hbn is now write enabled
Tue 09:41:34 /dev/rmt/1hbn mount operation in progress
Tue 09:41:50 /dev/rmt/1hbn mounted dlt7000 tape 000068
Tue 09:41:50 media event cleared: backup to pool 'Full' waiting for 1
writable backup tape(s)
Tue 09:41:50 beaker:<22>/raid/oscar50 saving to pool 'Full' (000068) 1766
MB
I've run all the diagnostic tests available on the front panel of the L9
and everything passes without problem. I can't find anything in the system
logs or the nsr logs to indicate what the problem might be. The E450 has
not been patched since it was installed. I plan to patch the machine this
week and I'm hoping there might be a patch to fix this, but I'm doubtful
since things appeared to be working fine until about Christmas time.
Things have slowly gotten worse over time to where I'm very worried about
the quality of our backups.
I'll be doing some testing to make sure that this problem is not
influencing the quality of our backups.
Any ideas what the problem might be?
Thanks in advance,
Stephen
_______________________________________________
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 - 23:28:10 EDT