Tim wrote:
> Take a look at the headers, as a message goes through a mail server, it
> adds its address above the last one. So the one closest to the bottom
> is the originator.
>
> However, be aware that a spammer may well send through a few servers
> before leaving out for the wide world. You can end up making a report
> directly to a spammer.
Remember rule one in spam-fighting: spammers *lie*. (And virus-writers
do, too). That means that you can't trust *any* headers for which they
might be responsible. In practice, that means you can't trust *any*
received headers that weren't added by a computer under your control or
your ISP's control.
Usually there'll be a line somewhat like
Received: from mail.ru ([80.197.104.24] RDNS failed) by smtp.example.com
(if your ISP is example.com) or
Received: from [85.140.207.188] (helo=menu2email.com) by smtp.example.com
You can't trust the *name* that the "from" computer reports, which in
these cases is mail.ru or menu2email.com. (Remember, spammers lie).
You probably *can* trust the IP address (80.197.104.24 or
85.140.207.188). This is because the e-mail travelled over TCP, and
unless someone managed to hack the routers between that computer and
you / your ISP, there was a conversation going on. smtp.example.com
sent packets to that address, and someone responded with the right
answers.
What you can tell from this is that the spam went through a computer at
that IP address. Either they're running an "open relay" (much less
common these days), which means they'll accept e-mail from anyone and
send it on to anyone else on the internet; or someone has "compromised"
that computer, or they're affiliated with the spammer.
Compromised routers are possible, but unusual -- insecure Windows
computers are a much softer target.
Hope this helps,
James.
--
E-mail: james@ | The TV networks don't know what to do about the loss of
aprilcottage.co.uk | eyeballs, and are prepared to try absolutely everything
| they can to get them back, apart from making stuff worth
| watching. -- Peter Corlett
