First suggestion, clean drives, at least twice. If tapes have ANY
write errors, make them candidates for immediate removal from your
pools. ... Yes it is expensive and frustrating, but it is all Ican
figure out what to do.
I have an older TSM on W2K with 3583-L72 with 6 LTO drives.
Sometimes it seems to help to shut down tsm server (power), turn off
the library (after all tapes are out of drives and back in their slots),
wait a while (some say sacrifice a chicken to the SCSI chain ;) during
this time) then power up the library. Let the library stabalize, then
bring the server up and online. ... nasty and time consuming, but it
helps to get rid of misc. issues. Usually before I shut down the library
I MANUALLY clean the drives and reset any error codes. Yes, I shouldn't do
it that way, but it is keeping my over burdended system running. (56 to 58
in the library tapedata tapes. I reclaim onsite at 50% just to keep it
barely working.
-----Original Message-----
From: Giedrius Jankauskas [mailto:Giedrius.Jankauskas AT MICROLINK DOT LT]
Sent: Friday, May 14, 2004 5:14 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Help needed ! Migration process fails :(
Hi there,
I have TSM (Version 5, Release 1, Level 5.0) installed on W2k SP4
server. Connected to it is IBM 3600-r20 library with ultrium td1 drive.
I'm having a problem which already gives me a headache. It's presented
in the activity log like this :
05/14/2004 12:50:49 ANR8337I LTO volume AA9004L1 mounted in drive
MT11.0.1.2
(mt11.0.1.2).
05/14/2004 12:51:00 ANR1340I Scratch volume AA9004L1 is now defined in
storage
pool ML_TAPEDAILYPOOL.
05/14/2004 12:54:05 ANR8311E An I/O error occurred while accessing
drive
MT11.0.1.2 (mt11.0.1.2) for WRITE operation,
errno =
1117.
05/14/2004 12:54:07 ANR1411W Access mode for volume AA9004L1 now set
to
"read-only" due to write error.
05/14/2004 12:54:07 ANR1181E astxn.c509: Data storage transaction
0:21378 was
aborted.
05/14/2004 12:54:07 ANR2183W dfmigr.c1965: Transaction 0:21378 was
aborted.
05/14/2004 12:54:07 ANR1033W Migration process 21 terminated for
storage pool
ML_DISKDAILYPOOL - transaction aborted.
05/14/2004 12:54:07 ANR0985I Process 21 for MIGRATION running in the
BACKGROUND completed with completion state
FAILURE at
12:54:07.
05/14/2004 12:54:07 ANR1002I Migration for storage pool
ML_DISKDAILYPOOL will
be retried in 60 seconds.
05/14/2004 12:54:12 ANR1341I Scratch volume AA9004L1 has been deleted
from
storage pool ML_TAPEDAILYPOOL.
It fails after writing solid amounts of data on tape (2-20GB).
The storage pool configuration is simple :
Storage Device Estimated Pct Pct High Low
Next
Pool Name Class Name Capacity Util Migr Mig Mig
Storage
(MB) Pct Pct
Pool
----------- ---------- ---------- ----- ----- ---- ---
-----------
DISKPOOL DISK 10,000.0 13.3 13.3 100 99
LTOPOOL1
LTOPOOL1 LTOCLASS1 0.0 0.0 0.0 90 70
ML_DISKARC- DISK 0.0 0.0 0.0 90 70
ML_TAPEARC
HIVEPOOL
HIVEPOOL
ML_DISKDAI- DISK 200,000.0 16.4 14.2 2 1
ML_TAPEDAI-
LYPOOL
LYPOOL
ML_TAPEARC- LTOCLASS1 0.0 0.0 0.0 90 70
HIVEPOOL
ML_TAPEDAI- LTOCLASS1 95,367,000 0.0 0.2 5 1
LYPOOL .0
SPACEMGPOOL DISK 0.0 0.0 0.0 90 70
If I backup or archive something directly to storage pool of ltoclass1
then it finishes without any erros, however the migration process always
failes. It then tries to mount another libvol and failes again... And so
on... I also successfully backed up files to tape with NTBackup. So this
happens only when I try to migrate the files :(
Does anyone have any ideas on why is this happening ? Haven't found
anything in the forums or mailing lists :(((
Grateful for any suggestions and will provide :((((
viny AT parkas DOT lt
|
