Unsolved
This post is more than 5 years old
1 Message
0
1221
November 3rd, 2010 08:00
Backups won't start and server_log contains NDMP_DEVICE_BUSY_ERR errors
Hi,
We're having an issue where backupexec tries to start backup jobs, but for some reason they stay queued forever.
Checking server_log, we saw that we have errors on the library:
Error opening SCSI device: c80t0l2 NDMP_DEVICE_BUSY_ERR
here's the output of server_devconfig:
jbox2 c80t0l2 jbox DELL PV-132T 310D310D GY004
It looks like restarting the server/tape library will get the job to start, but won't necessarily complete successfully. Everything worked fine up to now, nothing changed in the setup.
We use a NS480 with code 5.6.44-4
Anyone has a guess on what we should check?
thanks
0 events found
No Events found!
umichklewis_ac7b91
300 Posts
0
November 3rd, 2010 08:00
How is the robot control configured in BU? Are running BU 12.5 with Shared Storage? Does the backup fail at a media swap or some other point? Does BU have anything useful in the log (maybe a control timeout or some such)?
Thanks!
Karl
gbarretoxx1
2 Intern
•
366 Posts
0
November 7th, 2010 15:00
Hi,
are you using DDS ( Dynamic Drive Sharing ) ? If yes, set the param NDMP scsiReserve to "0". ( check with command server_param server_2 -f NDMP -i scsiReserve -v ). This might cause the device_busy errors.
Other possible cause is the drive index does not match the Celerra order. If this is the case, run a rescan ( server_devconfig server_2 -c -s -n ), and try again.
I hope it helps you.
Gustavo Barreto.
JGreen
46 Posts
0
November 8th, 2010 10:00
After the rescan make sure the backup software device configuration vs. Celerra tape device configuration matches before retesting.