RE: new deployment fails to receive email from "just" gmail successfully

2015-11-14 Thread Chris Boylan
Thank you, that was extremely useful information and explanation.

I made a new gmail test account and it worked fine out of gate.

Further research turned up that with gmail you can configure out-bound email
to go through a different mail server.  The original test account I used had
been configured that way to testing something with Reply-To and I wasn't aware
of it.  Sigh.

I'm doing the port 587 config next.

Thanks all for the help.


-Original Message-
From: owner-postfix-us...@postfix.org [mailto:owner-postfix-us...@postfix.org]
On Behalf Of Bill Cole
Sent: Friday, November 13, 2015 18:23
To: postfix-users@postfix.org
Cc: postfix-users@postfix.org
Subject: Re: new deployment fails to receive email from "just" gmail
successfully

On 13 Nov 2015, at 13:44, Chris Boylan wrote:

> There is no port number in the log messages you reference.  How do I 
> interpret
> what's there as a port 587 connection rather than a port 25 
> connection?

This log line:

> Nov 12 20:00:41 mail0 postfix/smtpd[24249]: send attr ident = 
> submission:209.85.160.172

Viktor's deeper understanding of Postfix debug lines confirms what I 
suspected but was unsure of, that 'submission' indicates that the 
connection is on the port used for message submission: 587. Note that 
you can avoid such uncertainty by overriding the syslog_name parameter 
for the submission service in master.cf, so that the submission and 
smtpd services tag their log lines differently.

Your 'postconf -M' output indicates that you have no config overrides 
for your submission service, so it will behave identically to your port 
25 smtp service. That is always wrong. For some unknown reason Google is 
trying to talk to you on the submission port and is dropping the 
connection AFTER establishing a TLS session and AFTER saying EHLO inside 
that session and AFTER you've responded to that EHLO. Were this a TLS 
problem, the failure would almost certainly NOT include any exchange of 
traffic inside the encrypted session.

If you were running a proper submission service on port 587, it would 
offer AUTH in its EHLO reply inside a TLS session, because a compliant 
MSA MUST (https://tools.ietf.org/html/rfc6409#section-4.3) require some 
form of authentication, normally the SMTP AUTH extension. If Google is 
expecting to see SMTP AUTH advertised in the EHLO reply and not seeing 
it (as they might, if you were a paying customer using the feature 
described at 
https://support.google.com/a/answer/2368131?hl=en&ref_topic=2921034 ) 
then a hard drop of the connection like that is not unreasonable.

The critical next step is to figure out WHY Google is using port 587 
instead of port 25. Since there are mandatory differences between SMTP 
and message submission that require senders to behave differently, it is 
reasonable to assume that when they decide for who-knows-what reason to 
use port 587, they also act differently than they would if they were 
using SMTP on port 25.



Re: Puting the Postfix's queue into RAM disk

2015-11-14 Thread Istvan Prosinger



On 13.11.2015 22:53, Phil Stracchino wrote:

On 11/13/15 14:17, Istvan Prosinger wrote:

I got two options that I know of. Signifficantly shortening the queue
lifetime, or (not) losing the queue from the RAM disk.
Just trying to measure which is worse (or to hear something new for me)


If you lack disk space for the queue, the server instance isn't up to
the job in the first place.

If you really think that keeping the queue in RAM is a better option
than on disk, despite that you probably have far more disk than RAM,
consider using a tmpfs instead of a RAMdisk, with the size of the tmpfs
capped at the size RAMdisk you were plannind.  A RAMdisk will use all of
that RAM all of the time, whether needed for queued mail or not.  A
tmpfs will consume only as much RAM at any given time as it needs right
then.





Thanks. It's not about a lack of space, puting it that word it's about a 
lack of an SSD drive.

Namely, it's about processing speed.