Re: [Exim] sender verify vs. broken mailer configs, again.

> > oh, you are so wrong here...! Reverse DNS is not a requirement of
> > SMTP, forward DNS is, and his forward DNS is fine.
>
> I never said Reverse DNS was a direct requirement of SMTP.
>
> However correct Reverse DNS is a requirement of the DNS.

But your SMTP server isn't doing that.

216.98.75.249 PTR is dial249.pm3abing3.abingdonpm.naxs.com
dial249.pm3abing3.abingdonpm.naxs.com A is 216.98.75.249

I have a hostname of animx.eu.org. A is also 216.98.75.249. What's the
problem?

> SMTP is very tightly intertwined with the DNS and thus correct DNS is na
> implicit requirement of fully functioning SMTP.

On the internet, yes. I can see SMTP not using DNS on a local LAN that is,
if it doesn't see the internet.

> However since reverse DNS isn't necessary when all else is correct you
> have two options: Use Reverse DNS correctly with correct SMTP
> configurations, or do not use Reverse DNS at all and hope your SMTP
> configuration is correct.

I've never had a problem with configuration. I'm not saying I don't check
some HELOs. I do do that, however, I'm not going as far as to verify that
the rDNS of every connecting system has to list the HELO name as one of the
PTRs.

I do check for HELOs that list
1) my IP
2) my domain
3) If the HELO is microsoft.com, compuserve.com, hotmail.com, msn.com,
aol.com, yahoo.com then I will at that time see if the rDNS contains the HELO
at the end. I do this because there are spammers that falsly do this.

#3 has never blocked legit mail.

> > and not accepting a sender callout attempt
> > a foolish act.
>
> No, what's foolish is software which allows an idiot administrator to
> mis-configure his system so that a sender callout attempt _must_ fail.

I did not do this. When you sent a message to my server, it attempted to do
a callout. It first said "EHLO animx.eu.org" which your server blocked and
requires HELO before MAIL FROM. This is NOT my misconfiguration.

> What would be smart would be software which detects when some remote
> client tells it that it is mis-configured and that if it is attempting a
> sender address verification then it would give up gracefully an inform
> its postmaster that there was a local configuration error.
>
> SMTP was not designed with sender callout attempts in question. If not
> very carefully implemented they cause deadlock situations. Exim's
> sender callout implementation is not careful enough.

Which is why callouts are done with a NULL sender.

--
Lab tests show that use of micro$oft causes cancer in lab animals