On Thu, Nov 26, 2015 at 12:03 PM, ale@proto <alessan...@protodigital.net> wrote:
> I reviewed my logs today and I saw a lot of connections from a bunch of MS > outbound gateways before entering the "postgrey layer". > > Once postscreen marked one of these gw PASS OLD postgrey put the message > in greylist (default 5 mins), but it expects another connection within > (better: after!) this time. This gw "disappeared" for 12 hours instead, > while another bunch of gateways hit my server. > > I know somebody discourages the use of postscreen + postgrey. But I don't > understand those MS retries. > > Here is my stripped log: > > Nov 24 17:51:13 MAILSERVER postfix/postscreen[21231]: CONNECT from > [157.55.234.104]:45788 to [MAILSERVER]:25 > Nov 24 17:51:20 MAILSERVER postfix/tlsproxy[21233]: CONNECT from > [157.55.234.104]:45788 > Nov 24 17:51:20 MAILSERVER postfix/postscreen[21231]: NOQUEUE: reject: > RCPT from [157.55.234.104]:45788: 450 4.3.2 Service currently > unavailable; from=<user@ms>, to=<recipient@here>, proto=ESMTP, helo=< > emea01-db3-obe.outbound.protection.outlook.com> > Nov 24 17:51:20 MAILSERVER postfix/tlsproxy[21233]: DISCONNECT > [157.55.234.104]:45788 > Nov 24 17:51:20 MAILSERVER postfix/postscreen[21231]: HANGUP after 0.21 > from [157.55.234.104]:45788 in tests after SMTP handshake > Nov 24 17:51:20 MAILSERVER postfix/postscreen[21231]: PASS NEW > [157.55.234.104]:45788 > Nov 24 17:51:20 MAILSERVER postfix/postscreen[21231]: DISCONNECT > [157.55.234.104]:45788 FYI - 157.55.234.0/24 is listed in the whitelist currently generated by Postwhite, so running it would probably get that mail to you many hours sooner.
