[milters] Archive

Lists Index Date Thread Search

Article: 3828
From: Anthony Howe
Date: 2014-01-09 13:28:48 -0500
Subject: Re: Strange error from milter-ahead

On 09 Jan 2014 12:54, Jon Rowlan shouted from a galaxy far far away...:
> Increasing verbosity of debug output I have found that this is caused
> by holding MX records in sendmail thus:
> 
> [x.x.x.x]:[y.y.y.y]

Huh?

> And when
> 
> [x.x.x.x] is offline, after 2 minutes of waiting for milter-ahead to
> connect to the lookup server, the connecting mail server gives up and
> I get:
> 
> EOM: unexpected close

The milter timeouts can be lengthened for the sendmail and the milter.

> I can get the IP's swapped but is there a way to avoid this? Perhaps
> a timeout setting that I can reduce?

Depends. There are the sendmail client timeouts (INPUT_MAIL_FILTER; see
Installation notes) and the milter timeouts in the milter-ahead.cf file
(see smtp-timeout, milter-timeout).

> As to my version, I am using a rather old Linux install and need to
> update, are the prerequisites changed and does my subscription
> include updates?

You need only update sendmail, libmilter, and the milter. BerkeleyDB
3.33 <= should be fine.

All buyers of a milter-NAME-1.x are entitled to any updates of the 1.x
series. Just login and download the latest 1.x (and related libsnert).

> I have relied on this software so much in the last decade or so that
> I would be happy to upgrade.

There is no milter-ahead 2.x upgades yet.

-- 
Anthony C Howe            Skype: SirWumpus                  SnertSoft
                        Twitter: SirWumpus      BarricadeMX & Milters
http://snert.com/      http://nanozen.info/     http://snertsoft.com/

Lists Index Date Thread Search