Upgrade to 7.6.0.2 job failures with Network errors 41 and 636

[cjm99]

After recent upgrade to 7.6.0.2 different job fail with network connection issues error 41 and 636.
I have looked through the forums and added the key to change the time out from 2 hours to 15 minutes.
The job complete when I retry on the 3rd time in the morning.

Job log below:

1st fail

10/07/2014 19:06:32 - Info bpbkar32(pid=8972) done. status: 41: network connection timed out     
10/07/2014 19:06:32 - end writing; write time: 1:02:40
network connection timed out(41)

2nd Fail

10/07/2014 19:36:58 - Info bptm(pid=1508) start           
10/07/2014 19:36:58 - begin writing
read from input socket failed(636)
10/07/2014 20:10:15 - Error bpbrm(pid=960) from client dcpwaexch03: ERR - Terminating backup.     
10/07/2014 20:10:15 - Error bpbrm(pid=960) from client dcpwaexch03: ERR - failure reading file: Microsoft Information Store:\DCPWAEXCH03 SG1\DCPWAEXCH03 SG1 DB1 (BEDS 0x0: )
10/07/2014 20:10:15 - Warning bpbrm(pid=960) from client dcpwaexch03: WRN - Exchange Validation for 'Microsoft Information Store:\DCPWAEXCH03 SG1\' failed.  Please refer to the backup and application event logs for more details.
10/07/2014 20:10:25 - Error bpbrm(pid=960) could not send server status message      
10/07/2014 20:10:27 - Error bpbrm(pid=960) could not write EXIT STATUS to OUTSOCK     
10/07/2014 20:10:27 - Info bpbkar32(pid=2896) done. status: 13: file read failed    

Status: read from input socket failed(636)

We never have issues with our network there are no firewalls between master and media servers. All servers are VM on VMware.

[NBU35]

you can try following dcoument if this is the case with only this client. (particular to exchange client)

http://www.symantec.com/business/support/index?page=content&id=TECH197008

also try to restart netbackup services on media & master servers.

[Mark_Solutions]

I have seen upgrades re-set your media server and client timeout values

It may be that you previosuly had them at high values but they are now lower so you are getting the timeout issues causing various job failures

Worth checking them all

[cjm99]

Hi,

The timeout values were standard values.

Logged with Symantec and the only suggestion is to change the  KeepAliveTime to 10 minutes from 15.

Not convinced this will work...

[Tanmoy1]

I'm not very sure though for this case but I've heard about some process instability of bpbkar in NBU 7.6.0.2 release apart from many other known issues. Though no official communication has got released yet I believe.Is this problem happening for a single client or multiple ?
 

[cjm99]

Update: Issue resolved.

On recommendation of support I changed the KeepAliveTime to 10 minutes and also updated the DataDomain OST plugin from version 2.6.0.2 to 2.6.2.0.

Had three days of error free backups.