NBU not attempting to write to new drives added

[screech27]

Hi, I hope I can get help.  I have an SR open, but so far no progress has been made.

We have a library with 6 drives - 2 LTO3, 2 LTO5 and 2 LTO6.  LTO3 use LTO3 tapes set up as hcart3.  LTO5 uses LTO5 tapes set up as hcart2.  We just added the 2 LTO6 drives.  Device monitor shows drives with the paths correct.  It is set as hcart2.  We use SLPs.  I changed the max concurrent drive in the Storage unit settings from 2 to 4.  We have duplicaiton jobs running - they use both the LTO5s and do not attempt to use the LTO6 drives.  message in policy showing it is waiting for drives is :awaiting resource oocsvopasb4-hcart2-robot-tld-1 Reason: Drives are in use, Media Server: oocsvopasb4"  I have checked, the concurrent drive for that SLP is set correct at 4.  We rebooted master and media server after the change, it didn't help.

I expect these LTO6 drives to use the LTO5 tapes - I know they cannot use the LTO3 tapes.  Not a problem there.

Any ideas?  We have never had a job run in these drives, they are new.  Thanks for any suggestions!

[Marianne]

Please show us output of tpconfig -l and vmoprcmd -d (Commands are in ...\volmgr\bin ) As well as output of bpstulist -U (in ....netbackup\bin\admincmd )

[mph999 · Accepted solution]

Editted my post as it wasn't correct - I'm on a different planet tonight ...

Further to Mariannes suggestion, if you 'down' the two LTO5 drives, leaving just the LTO6 drives up, does it then use them ?

[screech27]

Sorry for the delay.  After weeks of troubleshooting, it turned out all our problems were because of not rebooting our switches after making the changes in the configuration.  We found this out when we moved to a new switch with the same configuration.   All of a sudden the LTO5 and LTO6 drives were in the same groups and could communicate.

Thanks for all your help!

Sandy
 

[sdo]

Hi screech27 - I think the root cause of your problem is the fact that a SCSI reset was required. Certainly one thing I have alway done after adding new FC drives and presenting them to a server is... if they don't appear properly to a media server, then wait for a quiet moment on the media server, or schedule some done down time... And during the quiet time, to down any existing tape drives on the media server... Then reset the FC HBA port on the media server - which can be accomplished in either of two ways - use HBA Anyware or SAN surfer tools to reset the HBA ports - or to logon to the SAN switch and close the SAN switch port that is/are cables to the media server - wait 60 seconds - and then re-open the SAN switch port - and then back on the media server rescan for devices by getting the OS to refresh (easy in windows) or use something like a devfsadm command in Solaris - and then up the drives again - and then run reconfiguration wizard in NetBackup to create/update device configuration. The reason for resetting the HBA port or closing/re-opening the SAN switch port is... That those causes the HBA port WWN to have to re-login to the SAN fabric, which in turn causes a SCSI reset. If you google the terms PLogin and FLogin you will be able to learn what is actually going on. HTH.