[exim-dev] [Bug 1788] Deliveries to one ISP sometimes fail s…

Góra strony
Delete this message
Reply to this message
Autor: admin
Data:  
Dla: exim-dev
Temat: [exim-dev] [Bug 1788] Deliveries to one ISP sometimes fail silently after T=remote_smtp defer on first delivery attempt
https://bugs.exim.org/show_bug.cgi?id=1788

--- Comment #3 from Alex Presland <exim@???> ---
Hi Jeremy. Thanks for your prompt reply.

I'm unable to fully confirm as to whether the email went into the retry queue,
as the granularity of the munin graphing from a week ago doesn't let me see
that. Certainly, nothing is logged for this MessageID between the initial
delivery attempts and the "Completed".

Now that I've read the documentation that you linked (and if I've understood
the message correctly now), this post should probably be titled "Some
deliveries not re-attempted after 'T=remote_smtp defer (-53): retry time not
reached for any host' on first delivery attempt".

The name of the transport must have mislead me to think that it was actually
attempting the smtp delivery - oops!

I ran a "exim -bP | grep -i retry" and didn't get a configuration setting back
for retry_defer (as mentioned in the documentation that you linked), only the
following settings:
    dns_retry = 0
    retry_data_expire = 1w
    retry_interval_max = 1d


The problematic domains (plus the waitrose.com domain) all resolve to
mx.avasin.plus.net

With a "they're queued with other mails for the that MX's IP address(es)"
understanding from the documentation, I'm now looking into how this message
inter-relates with emails to the same destination MX. An initial look shows:
* Email 1aQLiq-0007QS-Q7 arrives at 21:09:20, successfully delivering to
user@??? via mx.avasin.plus.net at 21:09:24, and being
Complete.
* Then email 1aQLau-00078Q-An (which arrived at 21:01:08), was being delivered
to user@??? but deferred at 21:01:09 with "R=dnslookup
T=remote_smtp defer (-53): retry time not reached for any host"; was then
successfully delivered at 21:09:25 and being Complete.
* Then the Complete for 1aQLXt-0006xu-N5 (a different destination address
[user@???] at the same destination MX) is seen immediately
after in the logs.

The following 5 log lines for what I've just described are one after another in
the logs:

2016-02-01 21:09:24 1aQLiq-0007QS-Q7 => user@???
<alias123@???> R=dnslookup T=remote_smtp H=mx.avasin.plus.net
[212.159.9.200] C="250 D99N1s0040Nc4R90199QpY mail accepted for delivery"
2016-02-01 21:09:24 1aQLiq-0007QS-Q7 Completed
2016-02-01 21:09:25 1aQLau-00078Q-An => user@???
<alias123@???> R=dnslookup T=remote_smtp H=mx.avasin.plus.net
[212.159.9.200]* C="250 D99N1s0040Nc4R90199Rpb mail accepted for delivery"
2016-02-01 21:09:25 1aQLau-00078Q-An Completed
2016-02-01 21:09:25 1aQLXt-0006xu-N5 Completed

So, what does the * after the IP address in brackets mean? I'd expect to see a
re-delivery attempt logged for 1aQLXt-0006xu-N5 before 'Completed' is logged.
Would I be correct in assuming that all external delivery retry attempts will
be logged?

Hopefully this gives you better information than just the details of
1aQLXt-0006xu-N5 in isolation?

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