> > Reading between the lines of the error message, it looks to me like
> > meiway.com's mail server is programmed to take the MAIL FROM address,
> > connect back to the sending server
Wrong, but it's a fully understandable error when firing from the hip with
head up the ass.
The attempt to deliver is made against the @sender.domain's MX RDATA fields.
This is a perfectly valid prerequisite: "we´ll accept mail from the sender
if the @sender.domain's MX hosts accepts mail from us". Turn about's fair
play, mate. Not really very socio-pathic BOFH at all. It's actually of
kind a "Golden Rule Of SMTP". Receive from others as they receive from you.
>, and see if it can RCPT TO that
> > address before it will accept the incoming mail.
It's called "sender address verification" in postfix terms.
> > I have zero sympathy for that approach, but won't waste our list
> > bandwidth on exactly why it's a brain-dead sociopathic antispam
> > technique.
Excellent decision, and wrong "approach"! Opionated, self-righteous,
aggressive, half-cocked assholes waste enough bandwidth as it is.
>220 postgresql.org ESMTP Postfix (2.0.2)
>vrfy pgsql-admin-owner+M7187@postgresql.org
>252 pgsql-admin-owner+M7187@postgresql.org
>
>should have worked ... I know I was having problems with the use of the +
>in the email address, that I just fixed .. maybe this was just before that
>fix/
I will un-whitelist that ip and we'll see very quickly:
Feb 3 03:07:41 Mgw1 postfix/smtp[74990]: 929D1EF908:
to=<pgsql-admin-owner+M7227@postgresql.org>,
relay=mail.postgresql.org[64.49.215.8], delay=1, status=deliverable (250 Ok)
Excellent! Thanks for fixing postgres list server. It now validates like
every other discussion list server that we subscribe to, and as postgres's
list server did until it was very recently broken. sh!t happpens, end of
story.
Len
