06:57:36.696 [13465] <2> bpbrm sighandler: bpbrm timeout after 10800 seconds
06:57:36.696 [13465] <2> clear_held_signals: clearing signal mask stack, mask_stack_depth = 0
06:57:36.696 [13465] <2> bpbrm kill_child_process: start
06:57:36.697 [13465] <2> bpbrm wait_for_child: start
06:59:10.955 [13465] <2> bpbrm wait_for_child: child exit_status = 82 signal_status = 0
06:59:10.955 [13465] <2> inform_client_of_status: INF - Server status = 41
03:58:24.320 [13472] <2> bpbkar process_file: INF - /path/to/some/file is now size 12
03:58:24.320 [13472] <4> bpbkar PrintFile: /path/to/some/file
03:58:24.321 [13472] <2> bpbkar process_file: INF - /path/to/some/file is now size 12
03:58:24.322 [13472] <4> bpbkar PrintFile: /path/to/some/file
03:58:24.322 [13472] <2> bpbkar process_file: INF - /path/to/some/file is sparse: stat.st_size = 12, stat.st_blocks * 512 = 0
03:58:24.323 [13472] <2> bpbkar process_file: INF - /path/to/some/file is now size 12
From: Steve Cashman [mailto:[EMAIL PROTECTED]
Sent: Tuesday, May 09, 2006 2:11 PM
To: Justin Piszcz
Cc: Aaron Mills; veritas-bu@mailman.eng.auburn.edu
Subject: Re: [Veritas-bu] hanging bpbkar process
Often times the Media / Master server encounters and error and exits but cant notify the Client (they don't even attempt to) It sounds like you have at least some logs enabled on the Media server since you mention that you reviewed the Bpbrm log. So review the Media Server logs to see if you can get more information. What I often have seen is a network disconnect between the Client and the Media Server. This stops the backup as far as Bpbrm is concerned but Bpbkar does not know about it, he just keeps churning through data until he is done.
If this only happens on single client then it maybe something like a bad port, Nic, Driver ect. If it only happens on one Policy but other policies for that same client fail then you have a mystery on your hands. If you can post the relevant logs so we can browse them a bit (Bpbrm, Bpbkar, Bptm to start I would think)
mkdir /usr/openv/netbackup/logs/bpbkar on the client, add VERBOSE = 5
to the bp.conf and watch the logs, also you can make a lot of logging
directories on the server as well and tail them when the problematic
client is backing up/etc.
On 5/9/06, Aaron Mills <[EMAIL PROTECTED]> wrote:
>
>
> I had a backup timeout yesterday. bpbrm timed out after the configured
> interval (3 hours), but when I check to see what happened to the client
> process, bpbkar is still running (client/server on the same box) - it just
> hasn't done anything since three hours before the job timed out. The bpbkar
> log doesn't show anything useful. The process hums along and then just stops
> logging all of the sudden. This always seems to happen on the same job,
> though - never any others.
>
> Any ideas on where else I should look here?
>
> -Aaron
>
> Aaron Mills
> System Administrator
> Return Path, Inc.
> 303.642.4111
> [EMAIL PROTECTED]
> http://www.returnpath.biz
>
_______________________________________________
Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
