Re: [exim] Queueing-issue when greylisted....

Top Page
Delete this message
Reply to this message
Author: Patrick von der Hagen
Date:  
To: exim-users
Subject: Re: [exim] Queueing-issue when greylisted....
Philip Hazel schrieb:
> On Fri, 1 Sep 2006, Patrick von der Hagen wrote:
>
>> Imagine queue-runs every 15 minutes.
>>
>> 0m: Mail from a to b arrives and is greylisted =temporary error.
>> 14m: Mail from c to b arrives and is greylisted.
>
> Are you sure? With the default configuration, the fact that b had a
> temporary error should prevent further delivery attempts to b until its
> retry time has come. So if the retry time for b is 15m, there will be no
> delivery attempt for the second message when it arrives.

Sorry for answering so late, but I've been around in Germany and didn't
have a decent internet-connection...
So, basically I was quite sure, but when challenged by the ultimate
exim-guru, I immediately loose confidence. ;-)

Section 32 of the manual says: "When a delivery is not part of a queue
run (typically an immediate delivery on receipt of a message), the
routers are always run, and local deliveries are always attempted, even
if retry times are set for them. This makes for better behaviour if one
particular message is causing problems (for example, causing quota
overflow, or provoking an error in a filter file). If such a delivery
suffers a temporary failure, the retry data is updated as normal, and
subsequent delivery attempts from queue runs occur only when the retry
time for the local address is reached."

The incoming message is addressed to a local user, which has a
forward-file which forwards the message to newuser@???. The
way I interpreted the documentation, this causes an immediate delivery,
which fails and updates the retry data.

My retry-rules basically are "retry for four days, then quit." However,
I happen to find mail in my queue which has been in my queue for more
than eight days.


However, 32.10 might apply here, because there might be sender-addresses
which pass through the recipients greylisting, which messes with the
retry-data.

[...]
> This area of logic is notoriously complicated, so the only way to be
> absolutely sure would be be set up a test and check it out.

I'm currently implementing a new exim-heartbeat2-cluster, which will
give me lots of opportunities for testing those issues. ;-)

BTW, if anyone happens to have some OCF-compliant exim-scripts, I'd be
greatful if they were contributed to heartbeat2-project. Or at least
contributed to me. ;-)

-- 
CU,
    Patrick.