Mr David Robertson wrote:
> Many thanks.  You make some very interesting points.
> 
> I like the cut of your jib ;-)
> 
> You may have talked me into some more aggressive retry rules.
> 
> Regards
> 
> David

*trimmed*

All you have to do for a 'safe' experiment is:

- leave the final quit alone at first, adjust only the initial retry.

Rotate your logs.

- restart Eixm and run the new settings for an hour or a day.

Run eximstats on the old log, then on the new.

Voila!  It shows you the stats on deliveries and failures with their times.

Rotate the logs, try another setting.

I keep a couple of full year and multi-year logs for baseline.

Failure beyond ten minutes that retry will *ever* recover from is very 
much the exception. IOW - fallback retry boxen should not be needed ven 
IF your senders were zombots doing dictonery address attacks, ELSE MLM's 
using 'harvested' lists full of stale addresses [1].

Not a lot of servers are on twice-a-week dial-in these days, and those 
whose greylist want much over 5 minutes can safely be blacklisted to 
keep your queue clean.  Unless you are sending anime cartoons instead of 
text, their user-communities probably couldn't read the messages anyway.

;-)

Bill

[1] any decent MLM can either park or purge bounced addresses on its 
own, and according to its own settings for 'vacation' and such. That's 
why we use 'em.

-- 
## List details at http://lists.exim.org/mailman/listinfo/exim-users 
## Exim details at http://www.exim.org/
## Please use the Wiki with this list - http://wiki.exim.org/

Reply via email to