Duplicate job fails

Hello there,

I struggled with this for a while. Here is what I discovered:

If you have a duplication that fails, the next time your duplication runs it trys to get that nights back up plus the one that failed the night before (so it trys to dup two nights worth of backup - a good idea on Veritas' part assuming you have enough tape space to dup two+ nights of data). If you have a very small retention window for your backup-to-disk media set such that it gets overwritten right-away, the duplication job cannot find the media for the previous night(s) and hence fails with the error you are describing.

This is normally not a problem because most people keep more than a few days of backup-to-desk. But if your dup job fails two or three nights is a row (say your tape operator is on vacation or sick and there is no one on site to pop in a new tape each day) , it keeps trying to back up all the sets that is has missed (2 then 3 then 4 etc.) and eventually hits a day where the backup-to-disk has already be overwritten/erased and generates this error.

I have been told Veritas is going to fix this in a future release. In the mean time, I have found deleting both the back-to-disk and dup jobs and then recreating them clears out this counter and tells the dup job to backup just that nights backup.

Clear as mud?
Nelson

Thank you Nelson....

Since my policy contains overwrite protection period if one job fails I have to clean the media to overcome the overwrite protection period.......

Hopes the veritas techies have some more to say.....

Hi,

Please refer this technote:

http://support.veritas.com/docs/278589

NOTE : If we do not receive your reply within two business days, this post would be marked "assumed answered" and would be moved to "answered questions" pool.

Hi,

Sorry I forgot to mention. The tech note you specified is the first one I reffered when this error occured. But the error keeps coming again and again..Now I have made new policy and Backup to disk folders etc..Waiting to see what happens..

thank you

Hello Paul,

Do update us the result after recreating the policy.

Note : If we do not receive your reply within two business days, this post would be marked ‘assumed answered’ and would be moved to ‘answered questions’ pool.

Regards.

Hi,

Still the same. Getting the same error for duplicate of weekly full backup. Eventhough first 3 or 4 duplicates worked fine.

Hello,


Make sure that you have selected the specific source and destination device.

Are you duplicating the backup jobs from b2d to tape?

If yes, then have you cataloged the b2d before?

NOTE : If we do not receive your reply within two business days, this post would be marked assumed answered and would be moved to answered questions pool.

Hi,

Yes I am using b2d and duplicating from b2d to tape.
I have cataloged the folder once. Do I have to do it every time?..I mean if so I will have to schedule the job for catalog also..I don't think so.
Waiting to hear a solution.

thank you

Hi,

I am also getting the same error for my B2D2T policy. The whole jobs will run fine for some days and then the error start for the duplicate job. But I can see that the backup to disk job has been correctly write to the catalog as i can see them when I go to restore

Do let me know if anyone has the resolution to it (short of having to recreate the duplicate template in the policy)

Thanks

Is there a way to get out of this cycle? I have encountered a similar situation where a dup fails and the next dup is nearly twice as big as it should be. In my case my dup is done weekly and it is over 600GB. When it does this double-size behavior it runs through all of the tapes in my autoloader and throws off my tape rotation.

Is there a way to prevent this behavior?