[Mimedefang] Problems with Query.pm

[Brian Leyton]

I am attempting to move my sendmail/mimedefang installation over to a new
server.  The new one is running Fedora Core 5, and I am installing
everything from scratch using YUM (though I am copying over my old filters
and configuration).  I have hit a snag in my testing.  I'm sure I must be
missing something in my installation, but I can't figure out what it is.

Here are the messages I get from the sendmail log:

Sep  7 16:10:55 scratchy mimedefang-multiplexor[20885]: Starting slave 1
(pid 20944) (2 running): Bringing slaves up to minSlaves (2)
Sep  7 16:11:39 scratchy mimedefang-multiplexor[20885]: Slave 1 stderr: no
IP address given at /usr/lib/perl5/vendor_perl/5.8.8/Mail/SPF/Query.pm line
255, <STDIN> line 1.
Sep  7 16:11:39 scratchy mimedefang-multiplexor[20885]: Slave 1 died
prematurely -- check your filter rules
Sep  7 16:11:39 scratchy mimedefang[20900]: Error from multiplexor: ERR No
response from slave
Sep  7 16:11:39 scratchy sendmail[20945]: k87KBdpH020945: Milter:
from=<brian.leyton at gmail.com>, reject=451 4.3.2 Please try again later
Sep  7 16:11:39 scratchy mimedefang-multiplexor[20885]: Reap: slave 1 (pid
20944) exited normally with status 9 (SLAVE DIED UNEXPECTEDLY)
Sep  7 16:11:39 scratchy mimedefang-multiplexor[20885]: Slave 1 resource
usage: req=1, scans=0, user=1.592, sys=1.508, nswap=0, majflt=0,
minflt=8552, maxrss=0, bi=0, bo=0

When I look at the relevant section of query.pm, I see that it is exiting
because of a missing IP address (as it says in the error message):

  if (not ($query->{ipv4} and length $query->{ipv4})) {
    die "no IP address given";
  }

I just can't figure out how to fix this.  I ran a test on my filter and it
comes up clean.

Anyone have any ideas?

Brian Leyton
IT Manager
Commercial Petroleum Equipment