Melvyn Sopacua:
> On Saturday 31 January 2009 09:45:16 Wietse Venema wrote:
> > Melvyn Sopacua:
> > > On Saturday 31 January 2009 08:03:17 Wietse Venema wrote:
> > > > Melvyn Sopacua:
> > > > > The problem is that postfix will see "success" initially:
> > > > >
> > > > > - client sends
> > > > > - postfix receives
> > > > > - postfix relays, gets OK
> > > > > ----> postfix deletes queued message
> > > > > - exchange checks quota, decides mailbox is full
> > > > > - exchange sends DSN to client
> > > >
> > > > Use reject_unverified_recipient, with a persistent database. This
> > > > will check the Exchange response BEFORE accepting mail, and will
> > > > avoid hammering the Exchange server with repeated requests.
> > > >
> > > > http://www.postfix.org/ADDRESS_VERIFICATION_README.html
> > >
> > > VRFY i...@example.com
> > > 252 2.1.5 Cannot VRFY user, but will take message for <i...@example.com>
> >
> > POSTFIX DOES NOT USE THE VRFY COMMAND.
>
> Aha, actual probe messages. Still it does not solve my problem:
> Unfortunately, some major sites such as YAHOO do not reject unknown addresses
> in reply to the RCPT TO command, but report a delivery failure in response to
> end of DATA after a message is transferred. Postfix address verification does
> not work with such sites.
>
> Exchange works like YAHOO in this case. Mailbox restrictions are not checked
> during SMTP conversation but once the internal queue figured it out.
Well, that is just too bad. Postfix does not have to solve every
problem that happens in a down-stream MTA. We live in a free world,
and the market is always right.
Wietse
