status 636 / status 42 errors - backing up to MSDP pool

[lovethatcheese]

Hello. Running into some issues with a client, looking to possibly get some clarification on it....

background - Master / Media servers - Windows Server 2008 R2 Enterprise. Firewalls off, no anti-virus installed, UAC off. Client - Red Hat 6.2.

From the detailed status of the most recent 636 error:

1/3/2013 2:34:02 PM - Error bpbrm(pid=2936) could not write FILE ADDED message to OUTSOCK     
1/3/2013 2:34:10 PM - Error bpbrm(pid=2936) could not write FILE ADDED message to OUTSOCK     
1/3/2013 2:34:15 PM - Error bpbrm(pid=2936) could not write FILE ADDED message to OUTSOCK     
1/3/2013 2:34:20 PM - Error bpbrm(pid=2936) could not write FILE ADDED message to OUTSOCK     
1/3/2013 2:34:26 PM - Error bpbrm(pid=2936) could not write FILE ADDED message to OUTSOCK     
1/3/2013 2:34:31 PM - Info mumedia(pid=3328) StorageServer=PureDisk:mumedia; Report=PDDO Stats for (mumedia): scanned: 615471856 KB, CR sent: 3237918 KB, CR sent over FC: 0 KB, dedup: 99.5%
1/3/2013 2:34:32 PM - Info bpbrm(pid=2936) validating image for client mufs1        
read from input socket failed(636)

____

status 42:

 - begin writing
1/2/2013 9:40:54 PM - Error bpbrm(pid=4880) db_FLISTsend failed: network read failed (42)       
1/2/2013 9:41:13 PM - Info mumedia(pid=2864) StorageServer=PureDisk:mumedia; Report=PDDO Stats for (mumedia): scanned: 582591574 KB, CR sent: 133162 KB, CR sent over FC: 0 KB, dedup: 100.0%
1/3/2013 12:41:24 AM - end writing; write time: 03:21:22
network read failed(42)
1/3/2013 1:56:42 PM - job 840 was restarted as job 855

________

For status 42, I found this note and verified I have everything where it needs to be (and it was):

Resolution:
When the media server is trying to connect to the master server, the connection is being rejected by the master because the media server is not on the master server's list of media servers. It is also possible that the media server does not have an entry of itself in the server list on the media server.

To resolve this error, add the media server to the master server's media server list. Check the media server and verify the media server name is there on it's own server list.

__________

I've since had this client successfully run a differential job. Any ideas on why the above is happening?  I'd like to diagnose/fix any potential problem as I have duplicated these exact settings on 8 other media servers / clients. Any help is greatly appreciated.

-Scott

[lovethatcheese]

update - I'm getting this on another client now as well -

1/3/2013 6:00:00 PM - granted resource MediaID=@aaaah;DiskVolume=PureDiskVolume;DiskPool=apdp1;Path=PureDiskVolume;StorageServer=apnetbackup;MediaServer=apnetbackup
1/3/2013 6:00:00 PM - granted resource apsu1
1/3/2013 6:00:00 PM - estimated 1766 Kbytes needed
1/3/2013 6:00:00 PM - Info nbjm(pid=2092) started backup (backupid=apfs1_1357264800) job for client apfs1, policy apfs1, schedule Full on storage unit apsu1
1/3/2013 6:00:02 PM - started process bpbrm (544)
1/3/2013 6:00:03 PM - connecting
1/3/2013 6:00:05 PM - connected; connect time: 00:00:02
1/3/2013 6:00:12 PM - begin writing
read from input socket failed(636)

__________

However, the percent complete at the bottom shows 100% with 0 minutes remaining (and did show this before eventually throwing the status 636 - so the backup is completing, it's just not reporting the finished job?).

[Marianne]

The problem is not on client side - there is a comms issue between media server and master server (not name lookup, and not Server entries either, as other processes between master and media server are fine).
bpbrm on media server needs to update bpdbm on master server with catalog metadata (FILE ADDED message) which is not happening here. 

Two similar Connect discussions that was unfortunately never resolved.
I feel that suggestions by Mark_Solutions in both these posts are worth a try:

https://vox.veritas.com/t5/NetBackup/Error-40-on-my-media-server/m-p/456611

https://vox.veritas.com/t5/NetBackup/Backup-failing-with-status-40/m-p/384039
 

[Mark_Solutions]

Thnaks for the suggestions Marianne - my first thought was to add the file:

\netbackup\db\config\DPS_PROXYDEFAULTRECVTMO with a value of 800 in it to the Master and Media Servers

[TJ_Henning]

I recently resolved all of my Status 636 errors, and it was completely unrelated to Netbackup.  What we did was change the NIC order on the Media server, and that fixed the issue.  The actual LAN connection was at the bottom of the list below a couple of disabled NICs and the Data Domain.  After moving the active connection to the top, followed by the DD, all of the 636's dissipated.  We also checked the client server and validated the NIC orders as well. 

Windows 2003:

Open Control Panel \ Network Connections

Click "Advanced" then "Advanced Settings"

Rearrange the NIC order as desired

Windows 2008/2008R2:

Open Control Panel \ Network and Internet \ Network Connections

Press "ALT"

Click "Advanced" then "Advanced Settings"

Rearrange the NIC order as desired