I have more information on this problem.

I see these messages in the 'amdump' file:
  dumper: logfile.c@202: memory allocation failed (28 bytes requested)
  dumper: debug.c@345: memory allocation failed (22 bytes requested)

The 'debug.c' message is repeated many times.

I ran a periodic 'top' during last night's backup, and I found
that the dumper process got into a loop and quickly used up all
the memory.  Here is a report just before it died:

 PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
23204 amanda    25   0 2579M 2.5G   808 R    92.7 74.5   8:35 dumper

The start of the looping appears to have coincided with a tape
change.  Note that I use chg-zd-mtx and ammt.

Any ideas?

Thanks!

David

From: "John R. Jackson" <jrj AT purdue DOT edu>
Reply-To: jrj AT purdue DOT edu
To: "David Trusty" <dwtrusty AT hotmail DOT com>
CC: amanda-users AT amanda DOT org

Subject: Re: dumpers exiting with signal 11 in version 2.4.3? Date: Fri, 03 Jan 2003 18:22:19 -0500

>I received these messages at the end of a backup:
>
>  driver: dumper3 exited with signal 11

Oops.  :-)

>I am running version 2.4.3.

What OS?

>I believe most files were written to tape, but these messages look unusual.

>I did not find core files, but they may be somewhere I haven't yet
>discovered.

They should be in your Amanda debug directory, e.g. /tmp/amanda.  You
should be able to find out with:

  amadmin xx version | egrep 'AMANDA_DBGDIR|AMANDA_TMPDIR'

We really need to find them and run a debugger to get a traceback to
figure this out.

>David

John R. Jackson, Technical Software Specialist, jrj AT purdue DOT edu

_________________________________________________________________