Re: Mail in Submit Queue
- Bill Cole opined on Thursday 23-May-2013@15:36:24
> On 23 May 2013, at 13:51, LuKreme wrote:Heh. I’ll pass, I think.
>> On 22 May 2013, at 07:07 , "Bill Cole" <postfixlists-070913@...> wrote:
>> Yes, I see that *now*. Thanks. I think I was tunnel-visioned on the end of the line.
>> After looking up the original email I see this is a account that forwards mail to a gmail account, and gmail rejected the forwarded mail because it was spam.
>> /var/log/maillog.1.bz2:May 21 14:29:31 mail postfix/smtp: 3F635118B777: to=<*munged*@...>, orig_to=<*munged*>,
>> relay=gmail-smtp-in.l.google.com[220.127.116.11]:25, delay=1.1, delays=0.26/0.05/0.39/0.45, dsn=5.7.1, status=bounced (host gmail-smtp-in.l.google.com[18.104.22.168] said: 550-5.7.1 [22.214.171.124 12] Our system has detected that this message is 550-5.7.1 likely unsolicited mail. To reduce the amount of spam sent to Gmail, 550-5.7.1 this message has been blocked. Please visit 550-5.7.1 http://support.google.com/mail/bin/answer.py?hl=en&answer=188131 for 550 5.7.1 more information. cl1si4050394igc.54 - gsmtp (in reply to end of DATA command))
>> Short of not forwarding to gmail, anything I can do so that this results in dropping the mail (the 'wrong' thing) instead of trying to generate the appropriate bounce (the 'right' thing)?
> I think you could use sender_dependent_default_transport_maps to route bounces to a smtpd that uses nested_header_checks (ewww) to discard messages bearing the fingerprint of being the result of a forwarding attempt, such as a Delivered-To header containing a local mailbox.
> Implementation details are left as an exercise for the reader (bwahahaha...) Note that I have not done the exercise myself, so consider this a hand-wave in what seems to be the right direction, nothing more.
You’re the same Bill Cole who used to hang out with me on the SIMS list way back when Mac OS 8 was cutting edge, right?
Those were the days.
Sam, I thought I told you never to play--
- On 23 May 2013, at 22:18, LuKreme wrote:
> Bill Cole opined on Thursday 23-May-2013@15:36:24[...]
>> I think you could use sender_dependent_default_transport_maps toIt shouldn't be hard to do, at least not to the point of doing something
>> route bounces to a smtpd that uses nested_header_checks (ewww) to
>> discard messages
>> bearing the fingerprint of being the result of a forwarding attempt,
>> such as a Delivered-To header containing a local mailbox.
>> Implementation details are left as an exercise for the reader
>> (bwahahaha...) Note that I have not done the exercise myself, so
>> consider this a hand-wave in what seems to be the right direction,
>> nothing more.
> Heh. I’ll pass, I think.
reasonable 99% of the time. It could have a real long tail of edge cases
if you have lots of forwarding accounts getting legit messages which
tickle spam filters, but if you can accept dropping all bounces of
forwarded mail as a no-exceptions policy it would mostly be a project in
reading the docs.
> You’re the same Bill Cole who used to hang out with me on the SIMSCaught! I confess to having a diverse and disreputable history of
> list way back when Mac OS 8 was cutting edge, right?
bloviating on a wide range of mail software. And last week I had a
little hands-on with a SIMS system, helping to end its life mercifully.
> Those were the days.Simplicity in an MTA has great nostalgic appeal. On the other hand, at
that time I was also handcrafting sendmail.cf for sites that could not
be run on SIMS or anything else as simple. I much prefer wrangling
Postfix (and a little CGP, and the inescapable but these days always