ADSM-L

Re: Can SCSI error cause errors on LTO2 tapes?

2004-09-13 07:45:04
Subject: Re: Can SCSI error cause errors on LTO2 tapes?
From: Richard Sims <rbs AT BU DOT EDU>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Mon, 13 Sep 2004 07:44:30 -0400
Start with the obvious, and then work your way outward toward the
obscure...
As noted in previous postings, there were a lot of problems with
earlier 358x
microcode. Check the levels in your library and drives for currency.
Likewise
check that your device drivers are as current as your TSM server level.
If those are okay, isolate the problem area, which could be a single
drive, by
conducting controlled tests with only certain drives available. Review
your
historic TSM Activity Log for indications of questionable drive
reliability.
The brand or manufacturing batch of your tapes may also play a part.
FWIW, I haven't heard of any problems with the CM chip.
And, as in past postings, the type of restoral being performed may
influence
what's going on.

    Richard Sims

On Sep 13, 2004, at 3:00 AM, Bill Boyer wrote:

We have a situation where LTO2 tapes written in a 3583 library
successfully,
but will fail when it comes time to restore the data from them. When
the
tape is mounted for a restore operation..the icon on the 3583 drive
shows
that it is seeking, but it never goes anywhere. THe drive just sits
there.
No data is transferred. But yet when I do a move data off that
volumes, data
comes off. When the tape gets mounted for the restore, in the WIndows
event
log we see errors about "The device, \Device\Scsi\symmpi2, is not
ready for
access yet.". After this we have to halt the TSM server, reboot the
library,
unload the drives and reboot the TSM server. This happens in Windows
2000
and WIndows2003 at both TSM 5.1.7.1 and now at 5.2.3.0 server levels.
not
all tapes are doing this. Could that onboard chip in the LTO2 tapes be
causing problems? How would that have gotten corrupted?

Bill Boyer
"An Optimist is just a pessimist with no job experience."  - Scott
Adams

<Prev in Thread] Current Thread [Next in Thread>