bugtraq - Re: Why are postmasters distributing the MyDoom virus?

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <OF55DA5AD8.74D32313-ON88256E37.0005E272-88256E37.0006FF09@2roads.com>
Date: Tue, 10 Feb 2004 17:16:24 -0800
From: mgotts@...ads.com
To: dfs@...ringpenguin.com, rms@...puterbytesman.com
Cc: BUGTRAQ@...urityfocus.com
Subject: Re: Why are postmasters distributing the MyDoom virus?

> > I was looking over the MyDoom email messages that I received todayand 
found
> > about 15 copies of the worm which came from postmasters in bounce 
messages.
> > Some postmasters, when sending out a bounce message, include the 
original
> > email message as an attachment.  If a bounce message is for a
> > MyDoom-infected message, the bounce message will sometimes includean 
intact
> > copy of the MyDoom executable which can be run by mistake with a few 
mouse
> > clicks.
> 
> This is sometimes unavoidable.  A lot of MyDoom's go to nonexistent
> recipients, and when they are failed with a 5xx failure code, the 
sending
> relay (quite reasonably) includes the original message in the bounce.
> 

Not only do a lot of MyDoom's go to nonexistent recipients, they go to 
nonexistent recipients *by design*. These are intentional bounces where 
the worm is putting a simple but typically nonexistent email address in 
the "To:" and putting it's intended target address in the "From:". Then it 
uses the 5xx failures to deliver the worm via the bounce mechanism, so 
that the recipient gets an actual non-delivery report, not a faked one.

It is a subtle distinction, but I don't recall this method of delivery 
being used previously. All those bounces/rejections you see in your 
mail-server logs for "bill@...rdomain.com", "jane@...rdomain.com", 
"john@...rdomain.com", and "sam@...rdomain.com", etc. are NOT really 
attempts to deliver the worm to those made-up addresses (though I doubt 
the worm author would mind if the addresses did exist...). They are meant 
to bounce to the *actual* intended recipient, who is listed in the "From:"

I agree that the 5xx failures correctly bounce the complete message. But 
it is that predictable process that is being exploited by the worm author.

-- Mark