[Exim] Large message causing reset from foreign MTA

I am running exim 4.05 on SuSE Linux 7.3 as an inbound gateway aliasing mail
to a disperse set of exchange 5.5 servers.
Today I discovered a run of retry time not reached messages for one
particular exchange server from only one of my gateways. After tracing
through the retry database and a few tests, I discovered a message causing a
connection reset by peer. This message is a short bit of html and approx
200,000+ newlines.
I am personally convinced that exim is handling the message fine, since it
was sitting in the queue and being retried at our set intervals. I have
searched the web, particularly the MS knowledge base, the exim docs and
faqs, and this archive, but I can't find anything like this really. I work
with a crack team of MS admins, with me suggesting and setting up the exim
gateways. They are most assuredly point the finger at the gateway as the
problem.
I was hoping someone on this list may have a pointer to any known issues
like this one. I can tell you the exchange system is 5.5 SP4. It is running
some form of spam filtering software, not sure which this week. And, it is
sitting behind a PIX firewall. I was certain they had the fixup smtp
protocol enabled on it, but it now doesn't look like they do when I telnet
in to the smtp port.
I would just work through this with them to determine the nature, but they
are very stubborn and unwilling to offer any useful information. Any
pointers would be great. I kept the spool files if anyone wants a look.

TIA.
---
Peace be to you,
David Monk
david@???