marcinek
ADSM.ORG Member
Hi, I have TSM 5.5.2 running on W2003. Lib is 7 drive 3584. There is a 3 node GPFS cluster migrating lanfree some large multimedia files (about 10G each) running on SuSE SLES Linux boxes.
During the file recall some random parts of the files are makred as damaged, But when we run audit vol fix=y they becomes vaild again. There are no patterns about which file, drive, or cartrige will be involved. It seems that storage agent records incoretct metadata since files themselves are ok.
We tried all the stuff like changing zoning, move drives to separate from disks hba , updating firmware whenever it is applicable, upgrading tsm, sta, hsm and so on. the problem persisted even after moving from lto3 to lto4.
The questions are:
1. Does somebody already had this kind of problems?
2. How to debug STA? Is there any kind of activity log on STA?
3. Does validateprotocol=all work for lanfree nodes?
4. Any sugestions to try more?
During the file recall some random parts of the files are makred as damaged, But when we run audit vol fix=y they becomes vaild again. There are no patterns about which file, drive, or cartrige will be involved. It seems that storage agent records incoretct metadata since files themselves are ok.
We tried all the stuff like changing zoning, move drives to separate from disks hba , updating firmware whenever it is applicable, upgrading tsm, sta, hsm and so on. the problem persisted even after moving from lto3 to lto4.
The questions are:
1. Does somebody already had this kind of problems?
2. How to debug STA? Is there any kind of activity log on STA?
3. Does validateprotocol=all work for lanfree nodes?
4. Any sugestions to try more?