Do a robtest to make sure that there are not any tapes in the drives (usually happens when tapes are no changed correctly in the library)

Take a look in the Windows event logs - there should be entries in there that tell you exactly why the drives have gone down

When i tried to give the robtest on media server it shows me following message:

O:\Program Files\Veritas\Volmgr\bin>robtest
No locally-controlled robots with test utilities are configured

O:\Program Files\Veritas\Volmgr\bin>

Thanks,

Add VERBOSE entry to ...\veritas\volmgr\vm.conf on Media server.

Restart NBU Device Management Service on Media server.

UP the drive(s)

Next time a drive goes DOWN, the reason will be logged in Event Viewer Application log.

Hardware errors (e.g. HBA) will be logged in Event Viewer System log.

you need to run robtest on the robot control host - is the library shared? does another server control the robotics?

As of now we have following entry under vm.conf

O:\Program Files\Veritas\Volmgr>more vm.conf
MM_SERVER_NAME = Mediaserver.dadc.sbc.com

You really need to look at the Windows Event.log as Mark suggested. The event log will contain hints to why the drives are downed.

O:\Program Files\Veritas\Volmgr\bin>robtest
Configured robots with local control supporting test utilities:
  TLD(0)     robotic path = {6,0,1,1}

Robot Selection
---------------
  1)  TLD 0
  2)  none/quit
Enter choice: 1
1

Robot selected: TLD(0)   robotic path = {6,0,1,1}

Invoking robotic test utility:
O:\Program Files\Veritas\Volmgr\bin\tldtest.exe -rn 0 -r {6,0,1,1}

Opening {6,0,1,1}
MODE_SENSE complete
Enter tld commands (? returns help information)

Seems like robtest working from robot control host

so what does S D show?

As wrote here again and again, check Windows Event log(Application log and System log) first. when drive goes down, the reason must be logged.

if You can not find any relevant messages, then follow Marianne's suggustions - create debug log folder, add VERBOSE line into vm.conf, restart NetBackup, up drives, and reproduce this issue. 

Please perform below steps:

1. add VERBOSE into vm.conf file as suggested above;

2. Create folders: ltid, robots, reqlib and daemon

under directory: veritas/volmgr/debug folder

3. upped the drives and wait for the issue to come up again.

4. share the output of logs under above folders and on the event viewer.

Also share the results of s d command under robtest and try to move any tape into the drive.

Drv DriveName            Shared    Assigned        Comment
  0 T006_F1_S7_D7         No       

  1 T006_F2_S1_D10        No       -
  2 T006_F2_S2_D11        No       -
  3 T006_F2_S3_D12        No       -
  4 T006_F2_S4_D13        No       -
  5 T006_F2_S5_D14        No       -

I would like to add as index 0 drive is having one Frame1 and other 5 drives are configured on Frame2

On Frame1 drive 1-7 backups are running suceesful but on other 5 drives goes down which is configured on Frame2.

Does this make diffrence OR is there any problem with Frame 2 on library side?

We have added the VERBOSE entry and we found following logs for rest of 5 drives:

e.x-Operator/EMM server has DOWN'ed drive T006_F2_S3_D12 (device 3)

Thanks.

All drive are LTO Ultrium-2 and having same firmware version 73V1
And on Frame also same version - 8930