SLP tape dulication stuck

[Adnan_F]

Hello,

I've SLP jobs (duplication to tape) which appear to be stuck now for more than 35 hours increasing the queue for other SLP jobs.

Three drives in my environment, one of them seems to have some problem but other two drives are active however their job activity seems to be waiting for media.

The last line in both job activiteis is as follows:

11/28/2012 3:53:26 AM - Info bptm(pid=7040) INF - Waiting for positioning of media id ZxxxL5 on server <MediaServerName> for writing.

There is no error on job activity, however they seem to be stuck.

Any help would be much appreciated.

Regards,

Adnan F

[Mark_Solutions]

Is there an image ID in the overview tab of the job?

If not do you do GRT backups (Exchange, AD or Sharepoint)?

If the job is showing nothing it may well be a GRT backup that is duplicating

These grab the drive and hold it whilst it does the GRT processing which can take a long time

Ways to speed these uu are as follows (wont help with the current one though):

1. Disbale GRT duplication (once the image has left disk you will no longer be able to do GRT restores

2. Reduce your disk storage unit fragment size to a smaller value (say 5000MB)

3. Use a GRT Proxy Host (media server) or if you already do then try without using one to see which is best

If you decide to disable GRT during duplication (Master Servers Host Properties) then you need to change the setting and then try and kill the job off (not easy!). When it kicks back in it will not do the GRT part.

This all assumes my guess on what it could be

Hope this helps

[Adnan_F]

Hello Martin,

Appreciate your detailed response.

Yes there is an image ID in the job acitivity, it is a BMR duplication. Would it happen for BMR backups as well?

Regards,

Adnan

[Mark_Solutions]

No BMR are normal backups so it rules that out then.

It sounds like you have drive / robot issues

If possible cancel all jobs, make all SLPs inactive and then do a nbrbutil - resetALL to clear all allocation.

Sort you drives out (use robtest if needs be to make sure all drives are empty and make sure they are all UP) and then re-activate your SLPs

Also worth,once all jobs are cancelled, checking for oprhaned bptm and bpbrm processes running on your media severs

Hope this helps

[Adnan_F]

Thanks Mark for your help.

As it turned out, a tape cartridge was stuck in the drive. Netbackup didnt seem to catch that and hence I was assuming it got stuck. 

Had to remove the tape manually through the HP webconsole. Later deleted all backlogged SLP jobs and also checked out the nbrbutil command. Cleaned the drive and all is well now.