Actually, this isn’t a new tactic at all. There was a lot of this going on in the 1990s and early 2000s, but I haven’t seen such widespread use of fake Received headers for a while now. As mail is no longer relayed, what’s the point? And yet, it’s coming again. Take this recent example:
Received: from host101-187-static.229-95-b.business.telecomitalia.it (host101-187-static.229-95-b.business.telecomitalia.it [18.104.22.168])
by real-mail-server.example.com (8.14.4/8.14.4) with ESMTP id t8NAOpJS007947;
Wed, 23 Sep 2015 11:24:57 +0100 (BST)
Received: from remacdmzma03.rbs.com (mail09.rbs.com [22.214.171.124]) by mail.example.com (Postfix) with ESMTP id B849451943 for email@example.com; Wed, 23 Sep 2015 11:22:43 GMT)
Date: Wed, 23 Sep 2015 11:22:43 GMT
Thread-Topic: Emailing: bankfl.emt
From: "RBS" <firstname.lastname@example.org>
Subject: Bankline ROI - Password Re-activation Form
This is a multi-part message in MIME format.
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Please find the Re-activation form attached, send one per user ensuring only one box is selected in section 3. A signatory on the bank mandate must sign the form.
… etc …
Obviously the above has been re-written to use example.com, and the made-up-name was something random. The rest of the header is as it was. They’re obviously trying to convince you that your mail servers have already seen this this message, so it must be okay. This is such a dumb trick – does any spam filter bother to even look at earlier headers? Are they hoping that Bayesian analysis will score the incorrectly guessed mail server as particularly hammy?
But what’s doing this, and why? Is there a new spambot in town, or is there a new spam filter that’s susceptible to such a dumb trick?
As it stands, this was sent from a blacklisted IP address and the SPF fails for RBS anyway, and the English it was written by a virtual English illiterate. For what it’s worth, the payload was malware in a ZIP.