Duplicate Failure

I am trying to understand why my duplicate job failed this weekend. It dosent really give an error code that would explain it.

I have three policies created each with two templates. The three policies each backup a different set of servers to disk and are scheduled to start at the same time. The templates for each policy are the same. One is the configuration of the job to disk, the second is the duplicate of that disk job to tape once the backup to disk job completes. They are all configured the same way.

The three backup to disk jobs completed correctly, the first duplicate to disk job duplicated fine, the second duplicate job failed at 13gb of 211 gb written to tape but the third duplicate job ran successfully and copied all data like the first duplicate job. The jobs are set to append to tape for 3 days. So each duplicate job will write to the same tape until full and then span to the next.

What I don�t understand is why the first duplicate job completed successfully (98gb duplicated to tape, 98gb originally written to disk), the second duplicate job failed (13gb duplicated to tape but 211gb originally backed up to disk) then the third job started after this one and duplicated successfully (189gb to tape and 189gb written to disk) each job sent to the same tape, so its not a media issue, nor a drive issue as the third job finished just fine. The third job did NOT cause the second job to cancel either because it had failed duplicating to tape before the third job had finished writing to disk.

This is the error:

Job ended: Saturday, January 06, 2007 at 11:16:18 PM Completed status: Failed Final error: 0xe0008105 - Invalid Physical Volume Library Media Identifier. Final error category: Job Errors For additional information regarding this error refer to link V-79-57344-33029


Click an error below to locate it in the job log


Duplicate- Exch
Media mount failed.
Invalid Physical Volume Library Media Identifier.


I read the article but it dosent really relate. First of all I could understand maybe the tape needing to be cataloged if the other jobs failed but how could the first job write successfully, the second configured exactly the same fail after a few minutes and then third job complete successfully to the same media and the same drive?

Your assistance is appreciated since this makes no sense to me!

Thanks