Big problem.. help...

[Julian Field]

At 13:35 27/11/2002, you wrote:
>On Wed, 2002-11-27 at 06:18, Julian Field wrote:
> > These sorts of strange failures with big queues are often caused by you
> > running out of filehandles. See what "ulimit -a" says (if "ulimit -a" fails
> > then try "limit" as it varies between different shells), and try reducing
> > the maximum size of a batch to 50 or 100 and see if the same problem occurs
> > then.
> >
>
>I had already lowered the batch size to 100 based on your previous
>posts.  I could try lower I guess.
>
> > Max Unscanned Bytes Per Scan = 100000000
> > Max Unsafe Bytes Per Scan = 50000000
> > Max Unscanned Messages Per Scan = 100
> > Max Unsafe Messages Per Scan = 100
>
>ulimit -a
>
> > core file size        (blocks, -c) 0
> > data seg size         (kbytes, -d) unlimited
> > file size             (blocks, -f) unlimited
> > max locked memory     (kbytes, -l) unlimited
> > max memory size       (kbytes, -m) unlimited
> > open files                    (-n) 1024
> > pipe size          (512 bytes, -p) 8
> > stack size            (kbytes, -s) 8192
> > cpu time             (seconds, -t) unlimited
> > max user processes            (-u) 7168
> > virtual memory        (kbytes, -v) unlimited
>
>My whole setup is basically vanilla everything.  It's a dual PIII with 1
>gig of memory.  The number of files in the queue is not all that big at
>this point.  There are 924 total files (462 pairs) in the queue that
>won't allow MS to start up.  When I rename that directory back to
>mqueue.in, MailScanner won't start.  I'm becoming convinced that some
>particular mail file is causing the problem.  But, I'm no expert and
>could be wrong about that.  K12 is out for their thanksgiving break
>until next Monday so I'm not too concerned at the moment.  I intend to
>investigate in more detail this weekend to see if I can determine which
>file (or files) is causing the problem.  For now I'm going to have to
>put it on the shelf until then.  I'll post a followup if I come up with
>anything more definitive.

Try moving the qf+df pairs back in to the mqueue.in about 50 at a time, and
see if it stands that or whether there is 1 message that kills it. That way
you can get all but the troublesome message delivered.
--
Julian Field                Teaching Systems Manager
jkf at ecs.soton.ac.uk         Dept. of Electronics & Computer Science
Tel. 023 8059 2817          University of Southampton
                             Southampton SO17 1BJ