This is a multi-part message in MIME format.
--
Can anyone help with this exim 3 problem please?
I can't see anything obviously wrong with what he's trying to do, and
the behaviour he's getting isn't what I'd expect.
--
Reply-To: Tyson Whitehead <twhitehe@???>, 259857@???
Resent-From: Tyson Whitehead <twhitehe@???>
Original-Sender: Tyson Whitehead <tyson@???>
Resent-To: debian-bugs-dist@???
Resent-CC: Mark Baker <mbaker@???>
Resent-Date: Fri, 16 Jul 2004 23:18:02 UTC
Resent-Message-ID: <handler.259857.B.109001967627668@???>
From: Tyson Whitehead <twhitehe@???>
To: Debian Bug Tracking System <submit@???>
Date: Fri, 16 Jul 2004 19:12:45 -0400
Sender: Tyson Whitehead <tyson@???>
Delivered-To: submit@???
Resent-Sender: Debian BTS <debbugs@???>
Resent-Date: Sat, 17 Jul 2004 00:23:20 +0100
Subject: Bug#259857: exim: fallback_hosts fails if retry time is not reached on original hosts list
Package: exim
Version: 3.36-11
Severity: normal
I've configured my machine to deliver it's own mail if possible, and if
not to fall back on the organizations smtp server.
As a router configuration I have:
lookuphost:
driver = lookuphost
transport = remote_smtp
The remote_smtp transport is defined as (actual fallback name removed):
remote_smtp:
driver = smtp
fallback_hosts = smtp.blah.blah
This works fine most of the time. A typical delivery log in which
the fallback host is used is (names mangled) is:
-- 2004-07-16 15:58:39 1BlYqt-0001aj-00 <= twhitehe at uwo.ca
H=localhost [127.0.0.1] P=esmtp S=1933 id=2004 07161558.37311.twhitehe
at uwo.ca
-- 2004-07-16 15:58:49 1BlYqt-0001aj-00 toip4.blah.blah
[xxx.xxx.175.87]: No route to host
-- 2004-07-16 15:58:52 1BlYqt-0001aj-00 toip5.blah.blah
[xxx.xxx.175.88]: No route to host
-- 2004-07-16 15:58:55 1BlYqt-0001aj-00 toip3.blah.blah
[xxx.xxx.175.86]: No route to host
-- 2004-07-16 15:58:58 1BlYqt-0001aj-00 toip2.blah.blah
[xxx.xxx.175.85]: No route to host
-- 2004-07-16 15:59:43 1BlYqt-0001aj-00 toip1.blah.blah
[xxx.xxx.175.84]: No route to host
-- 2004-07-16 15:59:48 1BlYqt-0001aj-00 => person@???
R=lookuphost T=remote_smtp H=smtp.blah.blah [xxx.xxx.254.75]
-- 2004-07-16 15:59:48 1BlYqt-0001aj-00 Completed
But, suddenly, and seemingly randomly, delivering the exact same
person moments later, I get:
-- 2004-07-16 16:50:27 1BlZf1-0001cQ-00 <= twhitehe at uwo.ca
H=localhost [127.0.0.1] P=esmtp S=1420 id=2004 07161650.25231.twhitehe
at uwo.ca
-- 2004-07-16 16:50:28 1BlZf1-0001cQ-00 ** person@???
R=lookuphost T=remote_smtp: retry time not reached for any host after
a long failure period
-- 2004-07-16 16:50:29 1BlZf2-0001cS-00 <= <> R=1BlZf1-0001cQ-00
U=mail P=local S=2270
-- 2004-07-16 16:50:29 1BlZf1-0001cQ-00 Error message sent to twhitehe
at uwo.ca
-- 2004-07-16 16:50:29 1BlZf1-0001cQ-00 Completed
-- 2004-07-16 16:50:29 1BlZf2-0001cS-00 => twhitehe at uwo.ca
R=lookuphost T=remote_smtp H=smtp.blah.blah [ xxx.xxx.0.32]
-- 2004-07-16 16:50:29 1BlZf2-0001cS-00 Completed
I'm not sure, but I think the problem is that the fallback hosts for
the smtp transport is not used unless the retry time has been reached
for some of the original hosts (i.e. a bit of missed/bad logic around
the retry time host database and the fallback host for the smtp
transport).
This results in all but the first of rapid-fire emails failing, but
emails spread out over a longer period of time succeeding.
-T
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: alpha
Kernel: Linux 2.6.6
Locale: LANG=C, LC_CTYPE=C
Versions of packages exim depends on:
ii cron 3.0pl1-83 management of regular background p
ii libc6.1 2.3.2.ds1-13 GNU C Library: Shared libraries an
ii libdb3 3.2.9-20 Berkeley v3 Database Libraries [ru
ii libident 0.22-2.2 simple RFC1413 client library - ru
ii libldap2 2.1.30-2 OpenLDAP libraries
ii libpam0g 0.76-22 Pluggable Authentication Modules l
ii libpcre3 4.5-1.1 Perl 5 Compatible Regular Expressi
-- no debconf information
--
