On Thu 01 Jun 2006 at 15:26:40 +1000, Daniel Carosone wrote:
> Absolutely.  We expect there to be some fallout, and concrete
> constructive information from users about what breaks for them is
> valuable.

I haven't tried to convert my sendmail configuration (which isn't very
complex, I'd say) to postfix other than mentally (while reading through
al settings in main.cf), but here are the stumbling blocks I foresee:

- TLS. (At least the Postfix in -current seems to handle it, the one in
  3.0 doesn't)
- A central mailer setup, with sender domain rewriting (i.e. sender
  addresses like user@laptop.falu.nl converted into user@falu.nl) 
  ( FEATURE(`allmasquerade') FEATURE(`masquerade_envelope')
  FEATURE(`masquerade_entire_domain') ) and local delivery for the whole
  domain (i.e. for *.falu.nl). Despite documentation of "mydestination"
  speaking of "What domains to receive mail for" it seems it is about
  hostnames rather than domain names, and I want to (be able to) spefify
  whole domains (or at least one).
- milters: SpamAssassin/spamd in particular. I have seen web pages that
  seem to explain things, but they don't appear to have per-user configs
  and bouncing of spam with high scores at the DATA phase (i.e. before
  finally accepting it).
- Bouncing of mail with invalid Message-Id headers (I wrote a simple
  Header rule for it in sendmail:

    # Add a rule to check for valid message ids.
    # They must contain at least < @ . >

    HMessage-Id: $>CheckMessageId

    SCheckMessageId
    R< $+ @ $- . $+ >	$@ OK
    R$*			$#error $: 553 Header Error

- Use of blacklists like postmaster.rfc-ignorant.org to bounce mail at
  FROM phase (i.e. even before the DATA phase).

This may just be a lack of documentation in the obvious places, but
these issues remain questions after at least having fully read
/etc/postfix/main.cf and grepping though /usr/share/examples/postfix/*.

Maybe I'll think of more.

-Olaf.
-- 
___ Olaf 'Rhialto' Seibert      -- You author it, and I'll reader it.
\X/ rhialto/at/xs4all.nl        -- Cetero censeo "authored" delendum esse.