ADSM-L

Repeated mount of bad scratch tape

2015-10-04 17:25:43
Subject: Repeated mount of bad scratch tape
From: McClean, Frank [mailto:Frank_McClean AT CALPERS.CA DOT GOV]
To: ADSM-L AT VM.MARIST DOT EDU
Environment:Tsm server RS6000 S80
                        AIX 4.3.3.15
                        TSM 3.7.3.1
Tape library: 3494 with 3590E drives

Doing a backup of the onsite primary tapepool (collocated) to the
offsite
copy tapepool (non-collocated).
Job ran fine for several hours and used many scratch tapes.
Then suddenly!!!!!!
Got the following error:

12/25/00   03:34:53      ANR8337I 3590 volume NT1941 mounted in drive
RMT2

                          (/dev/rmt2).

12/25/00   03:34:53      ANR1340I Scratch volume NT1941 is now defined
in
storage
                          pool COPYPOOL.

12/25/00   03:34:56      ANR8302E I/O error on drive RMT2 (/dev/rmt2)
(OP=3DWRITE,
                          CC=3D0, KEY=3D03, ASC=3D85, ASCQ=3D00,


SENSE=3DF0.00.03.00.00.00.50.58.00.00.00.00.85.00.FE.0A.3A-

.5B.41.00.00.09.01.91.04.09.10.40.00.00.0A.00.00.33.3D.0-

0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.01.00.-
                          00.00.04.FF.00.00.00.00.03.00, =
Description=3DAn

                          undetermined error has occurred).  Refer to
Appendix B in
                          the 'Messages' manual for recommended action.

12/25/00   03:34:56      ANR8359E Media fault detected on 3590 volume
NT1941
in
                          drive RMT2 (/dev/rmt2) of library 3494LIB.

12/25/00   03:34:56      ANR1411W Access mode for volume NT1941 now set
to

                          "read-only" due to write error.

12/25/00   03:35:02      ANR1341I Scratch volume NT1941 has been =
deleted
from
                          storage pool COPYPOOL.

This sequence of messages repeated four times, same tape volume,
different
tape drives.  Then the job blew off:
12/25/00   03:58:46      ANR1219E BACKUP STGPOOL: Process 84 terminated
-
excessive
excessive
                          write errors encountered.

12/25/00   03:58:46      ANR0986I Process 84 for BACKUP STORAGE POOL
running
in the
                          BACKGROUND processed 909188 items for a total
of

                          144,400,306,523 bytes with a completion state
of
FAILURE
                          at 03:58:46.
NT1941 is the only scratch tape this process tried to mount after
receiving
the first error..
Is there a way to get TSM or the 3494 library manager to select a
different
scratch tape in the event of repeated failures????

Thanks in advance for any help, ideas, or suggestions.
Frank,

Frank McClean
Frank_McClean AT calpers.ca DOT gov