[exim] Problem with exim 4.63...

Hi all,

I have a odd problem right here on one of my servers. You must know that
this server never had problems, but suddenly I got a bogus of these
messages in paniclog:

1GDgZZ-000Ojz-Eq remote delivery process count got out of step
1GDgZZ-000Ojz-Eq remote delivery process count got out of step
1GDgZZ-000Ojz-Eq remote delivery process count got out of step
1GDgZZ-000Ojz-Eq remote delivery process count got out of step
1GDgZZ-000Ojz-Eq remote delivery process count got out of step
1GDgZZ-000Ojz-Eq remote delivery process count got out of step
1GDgZZ-000Ojz-Eq remote delivery process count got out of step
etc...

What happens you should ask? Well, my paniclog is flooding like hell, and
my diskspace is 100% within 10 minutes. When I restart exim, everything is
working fine again... But the funny thing is, this never happened before
and suddenly I have this problem. Alright, well, what is my OS... it's
FreeBSD 6.0. I have more than 20 servers with the same configuration and
and exim version 4.63 who does not have these problems, only one
machine...

So what I have done at the moment (because the machine is flipping like a
madddog) is writing a script like this (that is checking with cron every
minute):

#!/bin/sh

failed=`cat /var/log/exim/paniclog | grep "remote delivery process count
got out of step" | tail -10 | wc -l`

if [ $failed -gt 0 ]; then
/usr/local/etc/rc.d/exim restart
/bin/rm -fr /var/log/exim/mainlog
/bin/rm -fr /var/log/exim/paniclog
fi

Witch prevents me from checking the machine all the time... (and filling
up the space to 100%) but this is not a clean solution at all. It hapens
at random, no time given. Sometimes withing 6 hours, sometimes withing 12
hours, and sometimes within 10 minutes. So could this be a bug, and if so,
is there a workarround?

Thanks for reading.

Yours Faithfully,

Jim van Wel