Duplicate Job vs. Backup Job

[OS111112]

Hi,

I have a weekly backup job to tape; 559GB that took 21 hours to run (including verify).

I setup a duplication job for this job to duplicate to tape. At hour 26, this job is still has not finished copying to the tape and the verify still has not even started.

Shouldnt a tape to tape backup be quicker then a backup over the network to the tape?

As a test previous to this, I setup a duplication job from another backup set and it took half the time of the network backup. Why would this job be any different? When you setup the duplication job, if you chose between media or resource, is one faster than the other?

Ideas?

[Ken_Putnam]

I assume that the first backup was B2D?

how fragemented is the disk device?
is it a NAS or a local HD on the media server?

[OS111112]

Hi, - thanks for your reply!

No - the first backup is also backup to tape.

The thought to doing the duplication job is being that we do our weekly full backups to tape, instead of a separate regular 'pull from the network' job for monthly, if we just do a duplication of the last weekly full to the monthly tape the last weekend of the month, then we would save network traffic, and the duplication would work faster than pulling over the network. (That was a long sentence, did that make sense?)

For one job of 100GB, the backup from the network took 3 hours. Then as a test, I took that job and duplicated it from tape to tape and it took 1.5 hours.

Now I am doing it with one of last week's full jobs. The backup from the actual machine (network) took 21 hours. Now taking that job that is already on tape and copying it to another tape is already up to 30 hours - and still going.

[Ken_Putnam]

Same two tape devices as the test job?

what kind of CPU is BackupExec using?

I agree that this makes no sense at all

[OS111112]

Yes, same test devices as test job. I'm not sure about the CPU - the job has now finished (35 hours, in comparison to regular backup job that only took 22 hours..) but when I was looking at the config on the machine it was moving along just fine so I don't think that CPU or memory was pinned. If it was, I would have noticed the slugishness.

[shweta_rege]

Hello,


-- Did you select the option of verify after the backup?



-- Kindly try the backup deselecting that option..



Thank You,

Shweta

[OS111112]

Yes, the job is set to verify after the job. I don't want to turn it off becuase I want to make sure everything is successful.

I don't think that the verify portion of the job has anything to do with it, considering the job was still backing up in the amount of time that the original job took to backup AND verify.