I was thinking if the connection is using TLS, then connection caching
wouldn't work anyway. But it seems I was misreading that Postfix document,
it seems to say Postfix would not re-use the connection, but another MTA
with a different software architecture maybe could.
-----Original Message-----
From: George Skorup (Cyber Broadcasting) via Af
Sent: Wednesday, September 17, 2014 7:57 PM
To: af@afmug.com
Subject: Re: [AFMUG] TEST - IGNORE
Yes, TLS, and it has been this way since like.. ever. More years than I
can remember anyway. What is/was Beehive running on, Qmail? Looking
through my logs, coming from Beehive it was TLS as well. Coming from
WISPA is TLS, too. What MTA is WISPA running?
The Postfix TLS/connection caching thing could be the problem. Maybe
they're trying to hold the session open to deliver more messages? That
seems like a Postfix problem to me. But that doesn't make sense because
in all (normal) instances (WISPA, Beehive AF, Amazon AF), messages are
coming in for each recipient (me and Mike are on the same lists)
separately. Separate message ID and SMTP session for each recipient per
message. The only thing different with Amazon is they seem to "forget"
to send a QUIT once in a while which Exim on my end says violates RFC,
so whichever Amazon outbound MX it happens to be at the moment gets
connection rate limited for bad commands.
I see tons and tons of connection rate limiting 'because of notquit:
command-timeout' in my log all from obvious spam hosts, besides the
Amazon SES hosts anyway. So I do not want to ignore bad commands for
obvious reasons.
I'm not saying anyone is right or wrong. But other messages for other
Amazon services is doing the same thing. It's not just this list. It's
Amazon. Maybe they're dropping the connection early without sending a
QUIT because they are configured for impatience? I don't know.
On 9/17/2014 5:48 PM, Ken Hohhof via Af wrote:
George, are the SMTP connections to your mailserver using TLS? The reason
I ask has to do with SMTP connection caching.
http://www.postfix.org/CONNECTION_CACHE_README.html
-----Original Message----- From: George Skorup (Cyber Broadcasting) via Af
Sent: Wednesday, September 17, 2014 5:34 PM
To: af@afmug.com
Subject: Re: [AFMUG] TEST - IGNORE
In my case, it's not exactly greylisting, but something is happening on
Amazon's end that causing them to terminate SMTP sessions too early or
something like that. Overloading perhaps? Them not sending QUIT seems to
be the issue. RFCs say that's how you end a session. So whatever,
delayed. Meh. I'm not changing my end because we already get enough
crap. And because this is Exim & SpamAssassin on cPanel/WHM, almost
everything happens at the initial SMTP connection: RBL checks, host rate
limit checks, etc. As I said before, this wasn't happening when AF was
on Beehive's servers and it's not happening with the WISPA lists.
I'm not trying to be bitchy or nitpicky, just trying to help Paul and
Curtis understand what's happening from my POV.
On 9/17/2014 5:20 PM, Ken Hohhof via Af wrote:
Received here at 11:59:41.
Jay, I would look for something specific to your situation.
Either Amazon just doesn't like you, or if you are doing greylisting, I
would look at that as a possible cause.
Greylisting defers delivery on purpose, forces the sending MTA to queue
the message, and usually relies on the next attempt looking exactly the
same including source IP. Amazon may not use the same source IP, it may
come from a pool of MTAs. Big systems that do greylisting may
incorporate a mechanism to learn what IP ranges belong to the same
organization, but you may not have that level of sophistication.
Or probably I'm completely out in left field, but I can't think why I
would get every single email within a second or so, yet yours are being
delayed hours.
My mailserver is on a Rube Goldberg connection due to a previous storm
(it's at my house on an EoIP tunnel over AT&T U-Verse), so it's not like
I have a particularly speedy setup.
-----Original Message----- From: CBB - Jay Fuller via Af
Sent: Wednesday, September 17, 2014 4:12 PM
To: af@afmug.com
Subject: Re: [AFMUG] TEST - IGNORE
written 11:59 am , received 3:22 pm
(Fyi)
----- Original Message ----- From: Mathew Howard via Af
To: af@afmug.com
Sent: Wednesday, September 17, 2014 11:59 AM
Subject: Re: [AFMUG] TEST - IGNORE
I don't like fried pickles.
________________________________________
From: Af [af-boun...@afmug.com] on behalf of James Howard via Af
[af@afmug.com]
Sent: Wednesday, September 17, 2014 10:30 AM
To: af@afmug.com
Subject: Re: [AFMUG] TEST - IGNORE
It seems to be bordering on libelous for someone to accuse us of not
being able to stay on topic.
From: Af [mailto:af-boun...@afmug.com] On Behalf Of CBB - Jay Fuller via
Af
Sent: Wednesday, September 17, 2014 10:16 AM
To: af@afmug.com
Subject: Re: [AFMUG] TEST - IGNORE
mmmmm. fried pickles.
----- Original Message -----
From: Jay Weekley via Af
To: af@afmug.com<mailto:af@afmug.com>
Sent: Wednesday, September 17, 2014 10:08 AM
Subject: Re: [AFMUG] TEST - IGNORE
I like pickles. Does anyone else?
Adam Moffett via Af wrote:
>
> Test messages here never get ignored. Expect at least 4 replies saying
> whether you passed or failed and possibly a hijack that turns the test
> thread into some other conversation.
> (maybe this one is the hijack)
>
>>
>> Paul McCall, Pres.
>> PDMNet / Florida Broadband
>> 658 Old Dixie Highway
>> Vero Beach, FL 32962
>> 772-564-6800 office
>> 772-473-0352 cell
>> www.pdmnet.com<http://www.pdmnet.com/><http://www.pdmnet.com/%3e>
>>
pa...@pdmnet.net<mailto:pa...@pdmnet.net><mailto:pa...@pdmnet.net><mailto:pa...@pdmnet.net%3e>
>>
>
>
>
________________________________
Total Control Panel
Login<https://asp.reflexion.net/login?domain=litewire.net>
To:
ja...@litewire.net<https://asp.reflexion.net/address-properties?aID=242260993&domain=litewire.net>
From:
00000148842f45a9-e2dfdede-8f87-4a01-83b3-7d51bc808b65-000...@amazonses.com<https://asp.reflexion.net/address-properties?aID=2603669244&domain=litewire.net>
Message Score: 2
High (60): Pass
My Spam Blocking Level: High
Medium (75): Pass
Low (90): Pass
Block<https://asp.reflexion.net/FooterAction?ver=2&bl-sender-address=1&rID=242260993&aID=2603669244&domain=litewire.net>
this sender /
Block<https://asp.reflexion.net/FooterAction?ver=2&ent=1&bl-sender-address=1&rID=242260993&aID=2603669244&domain=litewire.net>
this sender enterprise-wide
Block<https://asp.reflexion.net/FooterAction?ver=2&bl-sender-domain=1&rID=242260993&aID=2603669244&domain=litewire.net>
amazonses.com /
Block<https://asp.reflexion.net/FooterAction?ver=2&ent=1&bl-sender-domain=1&rID=242260993&aID=2603669244&domain=litewire.net>
amazonses.com enterprise-wide
This message was delivered because the content filter score did not
exceed your filter level.
=
