One mail client's outgoing messages flagged as SPAM

Question

Hi again,
I have a new problem that has cropped up. Our company is multi-location. 8 of our locations are able to relay outgoing mail through our mail server with no problems whatsoever. Since upgrading clamav, amavisd-new and spamassassin (thanks topicdesk.com!), I have one remote client for whom every outgoing message is flagged as Spam by spamassassin and bounced. Here is the output from our mail server:
Content type: Spam
Internal reference code for the message is 05499-02/NFMMwedgHcWi
First upstream SMTP client IP address: [xx.xxx.xxx.153]
adsl-xx-xxx-xxx-153.dsl.lgtpmi.sbcglobal.net
According to a 'Received:' trace, the message apparently originated at:
[xx.xxx.xxx.153], adsl-xx-xxx-xxx-153.dsl.lgtpmi.sbcglobal.net
adsl-xx-xxx-xxx-153.dsl.lgtpmi.sbcglobal.net [xx.xxx.xxx.153]
Return-Path: <[email protected]>
From: Lansing <[email protected]>
Message-ID: <[email protected]>
X-Mailer: Apple Mail (2.1077)
Subject: test
The message has been quarantined as: spam-NFMMwedgHcWi.gz
The message WAS NOT relayed to:
<[email protected]>:
554 5.7.0 Reject, id=05499-02 - SPAM
SpamAssassin report:
Spam detection software, running on the system "server.stores", has
identified this incoming email as possible spam. The original message
has been attached to this so you can view it (if it isn't spam) or label
similar future email. If you have any questions, see
[email protected] for details.
Content preview: My Company of Lansing [email protected]
(Deleted to protect privacy)
Content analysis details: (8.2 points, 4.0 required)
pts rule name description
3.2 FHHELO_EQ_D_D_DD Helo is d-d-d-d
-1.9 BAYES_00 BODY: Bayes spam probability is 0 to 1%
[score: 0.0000]
1.0 RDNS_DYNAMIC Delivered to internal network by host with
dynamic-looking rDNS
0.2 HELODYNAMICDHCP Relay HELO'd using suspicious hostname (DHCP)
2.8 HELODYNAMICHCC Relay HELO'd using suspicious hostname (HCC)
1.0 TONO_BRKTSDYNIP TONO_BRKTSDYNIP
2.0 HELODYNAMICIPADDR Relay HELO'd using suspicious hostname (IP addr
1)
The only major difference in this location is that it's running a new Intel iMac with Snow Leopard. All the other clients are PPC's with Leopard.

Answer 1

As it turns out, the problem was with the router configuration. Instead of having the computer on a private network behind the router, the computer was given the public IP address. I'm not sure how this relates to how the message and its header are interpreted by spamassassin, but reconfiguring the network and using NAT to allow incoming traffic to the computer seems to have resolved the problem.
I would love for someone who knows about this stuff to respond, although technically my problem is resolved.
Thanks for reading!