Restore issue with media

[Raz_Sam]

Hi all,

I am facing restoration issue.

When I am trying to restore the data from tape, the restore job is in active state but, it is "awaiting resource". Even though the media is available in library.

This is found in job Detailed Status.

For this I have performed Robotic Inventory, still it is showing the same message.

When I have checked the media status in command line using "bpmedialist", it is showing Server Group=UNRESTRICTED_SHARING_GROUP.

For the above one, in Master server host properties, I have disabled, "Enable unrestricted media sharing for all media servers" option.

Still I am facing the same issue.

Can Any one suggest on this issue please, It would be helpful in resolving this issue.

Thanks in advance,

Sam.

[Marianne]

Please copy all text in Activity Monitor details tab and post here.

Also post output of this command for tape listed in restore job:

nbemmcmd -listmedia -mediaid <media-id>

[Shaf]

Sam,

If you change the disable "Enable unrestricted Media Sharing", I think it will effect to new backups not for existing backup,

please checking the following,

Check media reqired for the restore and then do filter in java gui--> media and check who is owning that tape...

and also check

FORCE_RESTORE option in bp.conf

[muhanad_daher]

try to disable the library driver and re-enable it, and restart the netbackup services, then start check inventory.

[Raz_Sam]

Hi Ah45,

Thanks for your reply. For your information, this is Windows platform. Can u plz suggest me how to check FORCE_RESTORE option in Windows machine.

Thanks in advance,

Sam

[Raz_Sam]

Hi Muhanad,

Please suggest me how to disable and re-enable the the library driver.

Thanks in advance,

Sam.

[Shaf]

Sam,

Check in this location

HKEY_LOCAL_MACHINE\Software\VERITAS\NetBackup\CurrentVersion\Config

for any FORCE_RESTORE_MEDIA_SERVER

[Marianne]

Please post output of 
nbemmcmd -listmedia -mediaid <media-id>
(command is in <install-path>\veritas\netbackup\bin\admincmd)

We are REALLY trying to help you, but we can only do so if you provide us with information...

You can check the GUI to see if FORCE_RESTORE_MEDIA_SERVER was enabled:
Host Properties -> Master server -> General Server -> Media Host Override

[Mark_Solutions]

As per my suggestions on your otehr thread:

Apart from the obvious things (tapes not in the library etc. which should all be shown in the text of the job itself) there are a couple of things that can cause this

1. Master Server Host Properties - Fibre transport - Default can be preferred - change to never of you dont use FT to speed things up

2. Master Server Host Properties - General Server - Delay on mulltiplexed restore - this setting allows you to kick off multiple restores before any will start to run so that they can do multiple restores from a multiplexed tape - however every restore from a multiplexed tape will be affected by this setting so if you do not often do multiple restores at the same time reduce this setting.

Changing these two can help - they do need a service re-start to register them

Hope these help

[bmathur]

As asked by Marrianne please send the output of nbemmcmd -listmedia -mediaid <media-id> and follow the instruction of Marrianne and Mark ,

if still the issue is not fixed then It may be an issue with the nbrb allocation , I have noticed it many times before to solve this issue please refer to the steps below.

1. nbrbutil - dump ( redirect the output to a text file)

2. Look for the media ID in the nbrb dump which is required for restore and see if its allocated and has an alloction key associated.

3. If there is a JOB actually running on that media server  then you will need to cancle that Job , cancelling a job is your decision if restore is critical you may want to cancle the job and run again. 

4.To relese the media from NBRB run the following command

 nbrbutil -releaseMDS <mediaID>

IF the media is still not deallocated then you may want to follow this technote

http://www.symantec.com/business/support/index?page=content&id=TECH126575

Cheers,

Boomchi

[Raz_Sam]

Hi Marianne,

Thanks for your help. Please find the attchements for output of "nbemmcmd" command and also detailed status of the job.

Another thing to inform you is, all the restores are facing the same issue.

Restore job is not getting failed, but it is in pending state for long time.

Please do needful.

Thanks in advance,

Sam.

[Raz_Sam]

Hi Marianne,

I have Checked,FORCE_RESTORE_MEDIA_SERVER is enabled.

Thanks,

Sam

[Marianne]

So, FORCE_RESTORE_MEDIA_SERVER is enabled. 
Care to tell us if this media server is involved? 
If so, does media server that needs to perform the restore have access to the robot where the tape is (robot 4)? And hcart drives?
Show us output of 'tpconfig -l' on the media server that needs to perform the restore.

Showing us nbemmcmd output with all hostnames blocked out does not help at all.

Please copy the text, then edit hostnames when you post it here in the forum, e.g:
Master
Media1
Media2, etc.

Then tell us which media server performed the backup and which media server needs to perform the restore.

Please also copy all text in Details tab of the restore job and post it here. You can once again edit hostnames.

For further troubleshooting, ensure that all of the following log folders exist:

On master: bprd (NBU needs to be restarted to enable this log. If Windows master, restart NBU Request Service)
On media server: bpbrm and bptm.
 

[Mark_Solutions]

I cannot really tell as you have XXX'd a lot of stuff out - but it looks like the tape is for robot 4 and the requested group is 00004 - this may be the same library but usually groups start at zero - that would indicate the request is for a different robot that the tape is in

It all depends on how you system is setup but it looks like the tape is in a different library to the request - and hence the server selected to do the restore does not have access to that library