Robert,
We have a L700 configured in a similar fashion except we have 8 SDLT
and 2 DLT drives and have seen some of the same problems. We were told
that SDLT did not need a cleaning tape. However, we have been seeing
lots of OS errors in the event log that these drives need to be cleaned.
Our backups have not been completing in the last week. Jobs will be
queue and even show what tape is being mounted even though the drive is
empty. I have seen yellow lights on the drives. Quantum originally
said they did not need cleaning tapes but now recommend using them when
the yellow lights are on the drives.
We purchased SLDT cleaning tapes from issibusiness.com yesterday for
$79.90. These are to be used for twenty cleans. Quantum now recommend
that when you see the yellow lights on the drive, you manually put in a
SDLT to clean it. The tapes arrived today and after we manually cleaned
the drives the lights went out.
When we had a 9710 library using DLT tapes only we set up cleaning
tapes to clean the drives after 20 hours. Since we have gone to the SSO
system with the L700 we manually need to run the tpclean -C drive name
command. If you type tpclean -l it will display the hours on the drive.
(install path\volmgr\bin) tpconfig -d will display the drive name along
with the index number and status of the drive. I have reset the time on
the drives by typing tpclean -M drive name.
If you only have one type of tape drive the L700 has a automatic clean
option that you can set up. Tapes go into the front door below the cap
or in a specified area and are labeled. You can set up the parameter of
how many hours between each clean and type of drive. Since we have a
mixed drive library I would rather manually clean these drives when the
error lights are displayed.
Hope this helps,
Kevin Wirick
Western Area Power Administration - DOE
Lakewood, CO
>>> Robert Johannes <ROBERT_JOHANNES AT udlp DOT com> 02/22/02 02:12PM >>>
I have 4 superdlt drives in an L700 configured for multihosted
environment (with SSO). The drives are connected via a scsi-to-fiber
bridge (crosspoint 4200, to be specific). As I understand it, SDLT
drives clean themselves, for that reason, I don't have cleaning tapes
in
the robot. I've observed that whenever nbu tries to clean the drives,
all the drives stop responding to nbu commands; thus backups or
restores
will sit, while netbackup waits for a response from the drives, which
never comes. The only way to recover from this is by entirely
rebooting
the machine.
Has anybody else experienced this behaviour? I figure if I can stop
nbu
from attempting to clean the drives all together, the drives then
wouldn't hung. Is there a bp.conf option, or some other option that
completely disables any sort of cleaning of the drives?
thanks
robert
_______________________________________________
Veritas-bu maillist - Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
|
