Chunk errors

[JG5]

I get random "Failed to copy or verify chunk" errors while doing aux copys. DNS is rock solid, all servers have plenty of power. It happens when I try to aux copy over my gig-e connection from our data center. Nearly every component has been replaced on the gig-e connection. I've noticed some CRC errors on that apparently are incoming on the gig-e port at the same time the aux copy begins. I'm going to try disabling flow control on the port to see if that makes a difference. Any suggestions?

Thanks.

 

[darren77uk]

I get the same deal, "sometimes" but I end up resetting the tape library and media agent and it seems to be fine. I have only come across this problem with the AIT 3 Drives, I have used CommVault at other places and on the LTO technology and it seems to work fine without any Chunk issues.

 

[JG5]

Mine is more constant than sporadic. It happens going from disk to disk as well. I've tried resetting all of the Media Agents also.

 

[BostonSteve]

I also get these constantly. I talked to support and they said that if you need to rely on this copy for restore you will not get any info that was in the chunk that failed. I have finally stopped relying on my aux copies. I am now relying on my Primaries for disaster recovery. When I do decide to do an aux copy I keep the copy on site and send the primary off site

 

[SwindoNut]

We have seen this chunk error also. Usually on aux copies. It has been suggested that the primary copy had a problem, although not picked up at the time.
Pause the aux copy, prune the primary job the error refers to, then resume job.
Of course, by doing this, the backup will be deleted, therefore its a good idea to run a fresh full backup.

 

[CVHack]

We have worked with CV support on many of these and most turn out to be bad primary copies, but CV support can't explain why we have a bad primary backup when it says successful!