[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

Alex Presland <exim@???> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|REMIND                      |---


--- Comment #28 from Alex Presland <exim@???> ---
Sorry that it has been ages since my last reply. The issue is still happening
on 'live' emails.

I've done three tests this evening, and proven that only one of them exhibits
this behaviour.

The tests were:
1) Email sent to two email addresses that both map in the domain alias file to
a single user account. The user account delivers both back to
"filtered-$original_local_part@$original_domain". One message is delivered.
The other is unable to be delivered at this time. Upon retry, the hints file
decides that the non-delivered message "was previously delivered: discarded".
SYMPTOMS OF THIS REPORTED ISSUE FOUND, in its simplest form.

2) Email sent to one email address that maps in the domain alias file to a
single user account. The user account delivers both this message to
"filtered-$original_local_part@$original_domain" and to an email address on an
external host. The email that goes back via the ailas file is delivered
without problem. The external host is unavailable, so the message is queued.
When retried, the message remains in the queue and/or is successfully delivered
when the remote host is back online. ISSUE *NOT* FOUND IN THIS SCENARIO.

3) Email sent to one email address that maps in the domain alias file to a
single user account. The user account delivered this message back to
"filtered-$original_local_part@$original_domain". The alias file takes this
email and sends it to two destinations. One is delivered without problem. The
second host is unavailable, so the message is queued. When retried, the
message remains in the queue and/or is successfully delivered when the remote
host is back online. ISSUE *NOT* FOUND IN THIS SCENARIO.


In each of the above 3 tests, a message was successfully delivered to the same
address(es) just before one of the destination hosts was taken offline (by
stopping the exim4 service). Once the destination host was made unavailable,
the test was performed.

For each test, once the message was in the queue it was manually processed with
a "exim4 -d+all -M <MessageID>" command. The following things have been logged
for each test (these details will be added as an attachment, in due course):
* The .forward file for the spamcatcher2 account
* The alias file for testdomain.org.uk
* The mailq, once every 60 seconds
* The output from the "exim4 -d+all -M <MessageID>"
* The mainlog details for <MessageID>


In conclusion, this bug remains valid and can be described as follows:
If an email is sent to multiple recipients, where both recipients have their
emails routed to the sameuser account by the alias file (and then back out via
the alias file), the Exim hints file will incorrectly cause one of these emails
to be discarded instead of being retried if the other was successfully
delivered during the initial processing.

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