I am indeed starting queue runners once an hour.
I'll search through the documentation a bit more, but in case someone can answer it off hand, does this mean that the first part of the retry rule stating every 15 minutes for 2 hours will never happen since a queue runner is not running within those intervals?
(btw, I did check about that invalid character, I think that was just some garbage left over from when I copied and pasted the information.)
If you look at the times listed on the log excerpts in my prior message, you will see that in the first example, the time elapsed between the original receipt of the local message and the total failure of delivery resulting in an error message was only two seconds. You can also see that the third MX record was never tried at all. This is what concerns me the most. I can live with the message hanging out in my queue for 5 hours until exim finally tried the third MX record, but I can't understand why it isn't being consistent either way.
Of course, I know there is always a reason for software being inconsistent, and it is almost always the fault of the person operating the software. :) I just need some help in figuring out where to look. I did a test run with: "exim -d9 -f xx@??? xx.xx@???" and that failed in the exact same way as the ones the log file recorded. There was no new information in the debug info. Just that it looked up the MX records for fedex.com, it found all three (with the multiple IP addresses for each) and tried all possible IP's for the first two MX's but not the third. :/ If anyone would like a log of the debug run, I'll be happy to post it. I just didn't want to spam the list. :) I looked at the information given by "exinext fedex.com" and it shows several failures that had been deferred and eventually failed entirely, but it does not show any of the failures that happen immediately. I might keep trying to resend a test message until I get exim to keep it in deferment so I can exinext it, but I'd like to investigate alternatives first.
Daniel
On Mon, 14 May 2001 09:12:50 +0100 (BST), Philip Hazel wrote:
On Sun, 13 May 2001, Daniel Einspanjer wrote:
> Retry rule: * F,2h,15m; G,16h,1h,1.5; F,4d,8h;
>
> Again, sometimes, it tries the first two MX records one right after
the other then immediately fails (without trying the third) and other
times, it will try the first two, wait an hour, try again, and after
several rounds of trying each hour on the hour, it will eventually try
the third and succeed.
How often do you start queue runners? It is once an hour?
The kind of behaviour you describe is perfectly possible, because Exim
will be keeping independent retry times for all three hosts. Therefore,
it is quite possible that for any particular delivery run, the retry
times for the first two hosts have arrived, but the third host's retry
time has not.
You can look at the retry information by using the exinext utility.
--
Philip Hazel University of Cambridge Computing Service,
ph10@??? Cambridge, England. Phone: +44 1223 334714.
--
## List details at
http://www.exim.org/mailman/listinfo/exim-users Exim details at
http://www.exim.org/ ##
