cancel
Showing results for 
Search instead for 
Did you mean: 

Error 636, multiple attempts and extra images

Genericus
Moderator
Moderator
   VIP   

Seeing some backups with multiple attempts, failing with error 636, then retrying and completeing successfully. All are VM backups with accelerator - here is the weird bit. The initial backups create a catalog image and are also processed by SLP.

This means that system runs a backup, when it completes it fails with a 636 and retries. However an image is created and the SLP runs, Second try, same thing. Third try succeeds. I have a successfull backup - that copies to tape three times.

NO big deal, for a 50GB backup, but when it is multiple TB, I really notice it!

I have found the TCP/IP timeout keepalive and will implement it next week, but thought it worthwhile to let you know to check the number of attempts and see if you have extra data out there!

 

NetBackup 9.1.0.1 on Solaris 11, writing to Data Domain 9800 7.7.4.0
duplicating via SLP to LTO5 & LTO8 in SL8500 via ACSLS
1 ACCEPTED SOLUTION

Accepted Solutions

Genericus
Moderator
Moderator
   VIP   

So, we discovered that we have two windows media servers, which we mount the VM disks on to do the backups, and someone had updated the policies to define a specific media server - when the data was mounted on media server 1, but it was being backed up to media server 2 storage, we had networking issues.

Once I hard coded the media server disk mount to match the storage, the issues went away...

 

NetBackup 9.1.0.1 on Solaris 11, writing to Data Domain 9800 7.7.4.0
duplicating via SLP to LTO5 & LTO8 in SL8500 via ACSLS

View solution in original post

3 REPLIES 3

RiaanBadenhorst
Moderator
Moderator
Partner    VIP    Accredited Certified

This sounds like some stange occurences where I've seen a backup complete, the image is created but some final comms between all parties involved fails (possible due to timeout) and the failure is registered, but the backup image is valid and you can restore from it.

The 636 would confirm it's network related so I'd say you've got a firewall or something else dropping connections, possibly between the master and media that cause the backup session to not close properly.

Mike_Gavrilov
Moderator
Moderator
Partner    VIP    Accredited Certified

I think it isn't weird. It just means that the network problem is between master and media, not the client and media.

 

Genericus
Moderator
Moderator
   VIP   

So, we discovered that we have two windows media servers, which we mount the VM disks on to do the backups, and someone had updated the policies to define a specific media server - when the data was mounted on media server 1, but it was being backed up to media server 2 storage, we had networking issues.

Once I hard coded the media server disk mount to match the storage, the issues went away...

 

NetBackup 9.1.0.1 on Solaris 11, writing to Data Domain 9800 7.7.4.0
duplicating via SLP to LTO5 & LTO8 in SL8500 via ACSLS