04-20-2017 07:58 AM
How does netbakup select the driver to use? We have 2 drivers for StorageTek SL150. But log show only one driver was used for different backups.
Solved! Go to Solution.
04-20-2017 11:52 PM
If configuration is correct and both drives are UP, NBU will automatically load balance the available tape drives.
Please show us config and drive status:
/usr/openv/netbackup/bin/admincmd/bpstulist -U
/usr/openv/volmgr/bin/tpconfig -l
/usr/openv/volmgr/bin/vmoprcmd -d
04-20-2017 08:05 AM - edited 04-20-2017 08:15 AM
There a couple of things to check:
For the storage unit, what is the value of "number of drives" set to ? (2 should be the value)
For the storage unit, what is the "max concurrent write drives" drives set to ? (2 shoud be the value).
04-20-2017 01:51 PM - edited 04-20-2017 01:52 PM
If we're talking about your case when check Nikolai's recommendations first.
If you just want to know how it works. Under the hood NetBackup has group of daemons (nbrb, mds, da, nbemm) sometimes called IRM (Intelligent Resource Manager). They monitor drives' state and select which drive should used for a job.
04-20-2017 11:52 PM
If configuration is correct and both drives are UP, NBU will automatically load balance the available tape drives.
Please show us config and drive status:
/usr/openv/netbackup/bin/admincmd/bpstulist -U
/usr/openv/volmgr/bin/tpconfig -l
/usr/openv/volmgr/bin/vmoprcmd -d
04-21-2017 08:26 AM
It has been my experience that the basic statement Marianne made - "If configuration is correct and both drives are UP, NBU will automatically load balance the available tape drives" is MOSTLY true (especially if storage unit is set to round robin)
NetBackup should take a bunch of other things into account -
are the drives shared? NetBackup will give priority to media server already allocated to minimize unload and loading
are multiple pools used? NetBackup will give priority to writing to pool already loaded.
With all other things equal - if you stop everything then start clean, it should load the drives alphabetically in order and round robin them - which when planning, means you can balance things like HBA and NIC using the drive naming...
04-23-2017 11:44 PM
Hi Marianne,
Thank you for the response. It show the drive was down. After bringing it up, both drives are now used.
Regards,
J
root@hq-vnb:~# /usr/openv/volmgr/bin/tpconfig -l Device Robot Drive Robot Drive Device Second Type Num Index Type DrNum Status Comment Name Path Device Path robot 0 - TLD - - - - /dev/sg/c0tw500104f000d23aa7l1 drive - 0 hcart3 1 DOWN - HP.ULTRIUM6-SCSI.000 /dev/rmt/0cbn drive - 1 hcart3 2 UP - HP.ULTRIUM6-SCSI.001 /dev/rmt/1cbn root@hq-vnb:~# root@hq-vnb:~# root@hq-vnb:~# /usr/openv/volmgr/bin/vmoprcmd -d PENDING REQUESTS <NONE> DRIVE STATUS Drv Type Control User Label RecMID ExtMID Ready Wr.Enbl. ReqId 0 hcart3 DOWN-TLD - No - 0 1 hcart3 TLD - No - 0 ADDITIONAL DRIVE STATUS Drv DriveName Shared Assigned Comment 0 HP.ULTRIUM6-SCSI.000 Yes - 1 HP.ULTRIUM6-SCSI.001 Yes -
root@hq-vnb:~# tpconfig -l Device Robot Drive Robot Drive Device Second Type Num Index Type DrNum Status Comment Name Path Device Path robot 0 - TLD - - - - /dev/sg/c0tw500104f000d23aa7l1 drive - 0 hcart3 1 UP - HP.ULTRIUM6-SCSI.000 /dev/rmt/0cbn drive - 1 hcart3 2 UP - HP.ULTRIUM6-SCSI.001 /dev/rmt/1cbn root@hq-vnb:~# vmoprcmd -d PENDING REQUESTS <NONE> DRIVE STATUS Drv Type Control User Label RecMID ExtMID Ready Wr.Enbl. ReqId 0 hcart3 TLD - No - 0 1 hcart3 TLD - No - 0 ADDITIONAL DRIVE STATUS Drv DriveName Shared Assigned Comment 0 HP.ULTRIUM6-SCSI.000 Yes - 1 HP.ULTRIUM6-SCSI.001 Yes -