> 1. What version of Exim are you using? Make sure you are using 4.68
> as there were some minor retry issues resolved in recent versions.
I'm using 4.50-8. I know it must be pretty old but I can't upgrade my
machines to a higher version right now. Too much trouble because I want
to stay with Debian packages. Upgrading exim would mean upgrading
the whole distribution and I'm not looking forward to much more serious
problems than exim retrying to deliver messages too frequently :)
> 2. What is the exact command line of the daemon that is running on
> your system (i.e. -bd -q15m, etc)?
exim -bd -q30m
> 2. Can you show the list the detailed log entries for one of the
> messages affected? Use exigrep so that you are sure to grab all the
> detail for that particular message id.
I'm attaching the relevant entries to this post. Since all
log entries are peppered with ip addresses, email addresses, etc,
I replaced all of that with the word "concealed" where appropriate.
Now, in the the attached log exim doesn't retry every 15 minutes. Instead it
seems like it goes from one MX IP to another round robin until finally
it delivers the message -- 5 hours later. I couldn't find a better example because the
queue is now empty. But I think this is actually what was happening
except I swear once in a while I would see 15 minute breaks between
retries.
> > Also does anybody know how to make exim stop trying to deliver a bounce
> > message? I would like one attempt made and if unsuccessful, the message
> > should be just discarded.
>
> One of the easiest ways would be to use a system_filter and inside it
> put something like:
>
> if error_message and not first_delivery
> then
> seen finish
> endif
Can the same thing be done with a rule?
