On Mon, 18 Mar 2002 10:33:12 GMT, Philip Hazel wrote:
> Messages don't fail. *Addresses* fail.
Yah, this is stuff I understand, I just didn't put enough effort into my
message. :-(
> So if you are talking about the "usual" modes of failure such as
> "connection refused", "connection timed out", etc., the retry time is
> measured from the first time delivery *to the host* failed.
Yup, that's what I'm talking about, and that's exactly the reassuring
answer I was looking for. :-)
> This is all documented in TFM.
Although the documentation is pretty thorough, Exim's quite a complex
beast (as is any reasonably featured MTA), and you shouldn't expect
folks to get a handle on it first time around every time. I read the
retry configuration section of the spec, and was reasonably sure that
retry times were relative to the time of first "failure", but if I'm
wrong, I'll end up in serious trouble with my boss.
All that said, my copy of the Exim book is finally on its way.
As always, thanks for the quick response, even in the face of
cluelessness. :-)
