Hi, I had the same problem, but was causing it myself in a Sun StorEdge
L1000 silo. Some backups were taking longer than normal and were not
complete when I would load the silo for the night's backups. If an active
backup was using a tape from a storage bin in which I had reloaded a tape,
when the drive went to return the tape, it found the bin full and 'downed'
the drive, even though there were other empty bins that it could have moved
the tape too. I finally caught on to this by watching in /var/adm/messages.
I now make sure that if a restore or backup is still running when I load, to
leave the slot empty where the active tape(s) came from.
From: Ravi Channavajhala [mailto:ravi.channavajhala AT csfb DOT com]
Sent: Friday, April 06, 2001 4:27 PM
To: veritas-bu AT mailman.eng.auburn DOT edu
Subject: [Veritas-bu] Drives going down randomly
Since the upgrade to Veritas DataCenter 3.4, I'm running into
an issue where the drives are going down very randomly.
Has any one experienced similar problems? The tape silo is
STK-9710, total 10 drives. I had STK replace the drives
which reported being 'down' and did all the necessary maintenance.
At this point, I like to see if any one has other suggestions?
I've tried virtually every trick in the book, so tpconfig etc.,
are already done. While we are at this topic, can anyone list
all the possible scenarios where NB software can down a perfectly
healthy drive, e.g., media mount problems etc., Thanks.
Veritas-bu maillist - Veritas-bu AT mailman.eng.auburn DOT edu