Video Screencast Help

My FS Bakup is failing with 25

Created: 10 Nov 2012 • Updated: 11 Nov 2012 | 8 comments
This issue has been solved. See solution.

Master Server 6.5.6 (Unix)

Media Server : 6.5.6 (Windows 2008)

Client :6.5.6 (Windows 2008)

Below is the error <16> write_backup: cannot add fragment to image database, error = cannot connect on socket

I have attached the client logs too...

Telnet/ping working good.....But when i go to client Host Properties getting error...Snapshot attached....Also bounced Media Server Services and Client ....

Any help will be appreciated.....

Discussion Filed Under:

Comments 8 CommentsJump to latest comment

revarooo's picture

Below is the error <16> write_backup: cannot add fragment to image database, error = cannot connect on socket

This means there is an issue with the media server sending the image data to bpdbm. Is this happening on all media servers or just one?
Check the Master server to see what the port usage is looking like:
netstat -a
see if there is any TIME_WAIT or CLOSE_WAITs perhaps?
2013's picture

Let me know the way how to check TIME_WAIT or CLOSE_WAITs and where i have to check....

RamNagalla's picture

hi SK,

as per my understanding you have a dedicated backup network (EBR), but it seemes you are not using the backup network in the Client side... does the client dont have the EBR interface?

below errors says its related to nework connection broken, using wrong interface also could cause this issue.

and you may need verify if its using the proper route to communicate with media server or not.. 

10:46:49.133 [37672.26260] <2> bpbrm main: client EXIT STATUS = 40: network connection broken
10:46:49.133 [37672.26260] <2> bpbrm kill_child_process_Ex: start
10:46:51.738 [37672.26260] <2> bpbrm wait_for_child: start
10:46:51.738 [37672.26260] <2> bpbrm wait_for_child: child exit_status = 25
10:46:51.738 [37672.26260] <2> inform_client_of_status: INF - Server status = 25
10:46:51.738 [37672.26260] <2> put_long: (11) network write() error: An existing connection was forcibly closed by the remote host. ; socket = 636
10:46:51.738 [37672.26260] <16> inform_client_of_status: could not send server status message

and related to Host properties incorrect passwd, check the below tech note


2013's picture

 I would like to highlight some more points...

1.On client EBR is not configured.

2.The Policy is backing up 3 Drives :F/G/H. Out of which F & G Drive backup completed successfully whereas the Backup fail on H Drive.

My question is if there is a network problem the F & G Drive backup should fail but they are completed successfully.

Need to know why only H Drive backup failed with 25 error?

Nov 11, 2012 11:03:24 AM - estimated 86094283 kbytes needed
Nov 11, 2012 11:03:26 AM - started process bpbrm (pid=36352)
Nov 11, 2012 11:03:27 AM - connecting
Nov 11, 2012 11:03:27 AM - connected; connect time: 0:00:00
Nov 11, 2012 11:03:36 AM - mounting AL4446
Nov 11, 2012 11:03:59 AM - mounted AL4446; mount time: 0:00:23
Nov 11, 2012 11:03:59 AM - positioning AL4446 to file 27
Nov 11, 2012 11:05:12 AM - positioned AL4446; position time: 0:01:13
Nov 11, 2012 11:05:12 AM - begin writing
Nov 11, 2012 11:33:07 AM - Error bptm (pid=22320) cannot add fragment to image database, error = cannot connect on socket
Nov 11, 2012 11:33:16 AM - Error bpbrm (pid=36352) could not send server status message
Nov 11, 2012 11:35:17 AM - Error bpbrm (pid=36352) cannot send mail because BPCD on exited with status 41: network connection timed out
Nov 11, 2012 11:35:18 AM - end writing; write time: 0:30:06
cannot connect on socket  (25)

revarooo's picture

Raise a tech support case with Symantec and they will ask you to run Apparanet networking tester between the client to media server. That may help.

Just because one drive will backup and another will not doesn't mean you don't have a network problem.

RamNagalla's picture

I agree with Revaroo..

we can not simply say it not a network problem, and even we can not confirm its a network issue... 

we have to look in deep before we come to one consulision..

Does H drive have the SQL database files?

did you keep the proper exclude list for not to backup SQL DB files directly...?

Try to increase Client read time out and Client connect time out on media servers  and check...

and also check how many hops its taking while communicating from Client to media server (long route also can cause this issue)

I understand who is your client, anything is possible with your client, we need to look in 360 degrees view to get this issue fixed..

Good luck.. let me know if you have any questions... 

2013's picture

Issue is now resolved after increasing "Client read time out and Client connect time out on media servers"

RamNagalla's picture

thats a good news..

it seems its because of the files its reading...may be SQL DB files.. 

make sure you have proper exclude list in place... and make sure its not  taking the SQL DB files