[exim-dev] [Bug 1788] Some deliveries not re-attempted after…

Top Page
Delete this message
Reply to this message
Author: admin
Date:  
To: exim-dev
Old-Topics: [exim-dev] [Bug 1788] New: Deliveries to one ISP sometimes fail silently after T=remote_smtp defer on first delivery attempt
Subject: [exim-dev] [Bug 1788] Some deliveries not re-attempted after 'T=remote_smtp defer (-53): retry time not reached for any host' on first delivery attempt
https://bugs.exim.org/show_bug.cgi?id=1788

--- Comment #13 from Alex Presland <exim@???> ---
Good news! I have managed to recreate the problem, while I have a tcpdump
running.

I did this by sending multiple emails to the user@???
address, until it I got a message back saying:
2016-02-29 21:24:09 1aaVIT-0006zh-Bo == user@??? R=dnslookup
T=remote_smtp defer (0): SMTP error from remote mail server after initial
connection: host mx.avasin.plus.net [212.159.9.200]: 421 avasin10.plus.net
Plusnet Cloudmark Gateway Too many connections from same IP

I then sent in a pre-prepared email to a normal mailman mailing list that
included:
- Two email addresses under in the userdomain.idps.co.uk domain
- One email address under the userdomain.plus.com domain
- One email address under the waitrose.com domain
(all of these domains end up at PlusNet's MXs with IP addresses 212.159.9.200
and 212.159.8.200)

The email to user1@??? and user2@??? were
presented to the server by mailman in one email [1aaVJn-0007Aa-9q]; and the
email to the user@??? and user@??? were presented to
the server by mailman in a second email [1aaVJn-0007Aa-0b]. Mailman presented
a third email for two other user domains [1aaVJm-0007Aa-K3].

I'll attach the logs for all relevant messages. One relevant line which wasn't
picked up by the grep on the messageIDs was:
2016-02-29 21:30:33 Start queue run: pid=28458

The tcpdump that was running in the background was called with the command:
tcpdump -n "(host 212.159.8.200 or host 212.159.9.200)" -C 100 -W 10 -w
/tmp/eximbug1788.cap &

My script that analysed the exim logs found a "missing delivery" at 21:31:27
and ran a "killall tcpdump" to stop the capture.

The two emails that went missing were for user1@??? and
user2@???. These were originally submitted by mailman to
first1.last1@??? and first2.last2@???. The alias file
for domain1.org.uk directed them to user accounts localuser1 and localuser2.
The .forward in each of these accounts caused the message to be delivered
locally (R=local_user T=maildir_home) and then used the eximfilter "deliver"
directive to deliver a copy to ...@userdomain.idps.co.uk.

--
You are receiving this mail because:
You are on the CC list for the bug.