<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=us-ascii">
<META content="MSHTML 6.00.6000.16608" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT face=Arial size=2>
<DIV class=headline> </DIV>
<DIV class=headline><A href="http://isc.sans.org/diary.html?storyid=4834">When
spammers use your own e-mails</A> </DIV>
<DIV class=diaryheader>Published: 2008-08-06,<BR>Last Updated: 2008-08-06
12:49:47 UTC<BR>by Bojan Zdrnja (Version: 1) </DIV><A
href="http://isc.sans.org/diary.html?storyid=4834#comment">0 comment(s)</A>
<SCRIPT type=text/javascript>
digg_url = 'http://isc.sans.org/diary.html?storyid=4834&rss';
digg_title = 'When spammers use your own e-mails';
digg_skin='compact';
digg_topic = 'security';
</SCRIPT>
<SCRIPT src="http://digg.com/tools/diggthis.js" type=text/javascript></SCRIPT>
<IFRAME
src="http://digg.com/tools/diggthis.php?u=http%3A//isc.sans.org/diary.html%3Fstoryid%3D4834%26rss&t=When%20spammers%20use%20your%20own%20e-mails&c=security&s=compact"
frameBorder=0 width=120 scrolling=no height=18></IFRAME>
<DIV class=diarybody>
<P>Some time ago, one of our readers, Mike S, sent an e-mail with an interesting
observation about how spammers used e-mails from one of his customers (this has
been actually sitting in my own inbox for way too long).<BR><BR>The e-mails
contained all "standard" elements such as spoofed headers etc, but there was a
very interesting thing with the body content.<BR><BR>As with most e-mail
spammers send, these e-mails were HTML as well. However, the interesting part
was that the spammers took his clients' e-mails and modified the HTML a bit to
include their own message.<BR><BR>The spammers added the link they wanted to
spam at the top and then opened a <TITLE> HTML tag. After the TITLE tag
came the full original e-mail, but the tag was never actually closed. This
resulted in Outlook displaying only the spammed link, but not showing the
original e-mail content. <BR><BR>The raw e-mail looked like this:<BR><BR><SPAN
style="FONT-FAMILY: Courier New">--AlternativeBoundary.22222222.22222222<BR>Content-Type:
text/html; charset="ISO-8859-1"<BR>Content-Transfer-Encoding:
8bit<BR><BR><html><center><FONT SIZE="5"
COLOR="#10566D">Spammers message</font><br><br><A
HREF="http://spammers link">http://spammers
link</A><BR><title><body leftmargin=5 topmargin=5 marginwidth=0
marginheight=0><BR><table width=100% cellpadding=0 cellspacing=0
bgcolor=white align=center border=0><BR><tr><td style='{font-family:
Verdana, sans-serif; color=#7a929f;font-weight:700;font-size:
11px;text-transform : capitalize;}'><BR>.... ORIGINAL MAIL CONTENT ...
<BR></td></tr><BR></table><p>&nbsp;</p><BR></body></SPAN><BR><BR>Of
course, by using the original e-mail content (which was legitimate when the
client sent it), the spammers are trying to evade Bayesian filters, and at least
in Mike's example they even managed to get SpamAssassin decrease the final score
of the e-mail.<BR><BR>In any case, it's an arms race between spammers and
content filter developers. Thanks Mike again for sending this interesting
information (and sorry it took so long to analyze
it).<BR><BR>--<BR>Bojan</P></DIV></FONT></DIV>
<DIV><A href="http://isc.sans.org/diary.html?storyid=4834"><FONT
face=Arial><FONT size=2><SPAN class=960385613-06082008><FONT
color=#000000>Source:
</FONT></SPAN>http://isc.sans.org/diary.html?storyid=4834</FONT></FONT></A></DIV></BODY></HTML>