[ On Monday, October 18, 2004 at 10:05:11 (+0100), Ian Eiloart wrote: ]
> Subject: Re: [exim] Is there and logical reason to reject mail from: <> ?
>
> Nobody is talking about ONE reason. THREE conditions have to be matched,
> afaict:
>
> 1. The email is addressed to POSTMASTER (a commonly forged email address).
> 2. POSTMASTER is an address that doesn't send mail.
> 3. The email has a null sender address - and is, therefore, an automatic
> reply.
That's not really three conditions -- that's really just the one
condition we've been discussing all along, or maybe one and a half. :-)
(seriously, an additional condition might be that the source address, or
a hostname, of the incoming connection matches some DNSBL, or the
content contains a forbidden type of content, etc.)
> A little bit of logic tells us that the email must therefore be
> backscatter, a callout, or a manual test of whether I accept email to
> postmaster from the null sender.
Yes, exactly.
> The RFCs tell me to accept email addressed to postmaster such that people
> can easily contact me. Provided I don't do callouts on email addressed to
> postmaster, rejecting backscatter shouldn't hurt there.
It's not your callouts that would be at issue, and you would not be
rejecting just and only backscatter.
As I've been saying -- there's no good reason for using this kind of
lame sledge hammer trick to try to put a dent in one tiny little itty
bitty bit of the MUCH bigger problem of backscatter. Learn to use the
right tool(s) for the job!
--
Greg A. Woods
+1 416 218-0098 VE3TCP RoboHack <woods@???>
Planix, Inc. <woods@???> Secrets of the Weird <woods@???>
