On 27 Oct 2012, at 12:30, [email protected] wrote:

> Yes:

That seems to be a red herring. Such a (temporary) failure as not being able to 
communicate with one of several name servers should never result in a hard 5xx 
error. Especially when it involves accepting to a local domain. 

That DNS failure of the sending host  coinciding with local domain rejections 
seems to be a symptom of whatever is actually failing, rather than the cause of 
the failure.

I did some subsequent testing with a separate exim process started on a 
different port. It hadn't processed any real SMTP traffic, and could only be 
connected to from my internal network. I got the same error after it had been 
running for a period of time (but, a "live" exim had been running and once 
again failed. So it's also possible that there's some corruption being caused 
somewhere by whatever real-world emails it was accepting)

Of course now that I've built a binary with debugging symbols - Murphy's law - 
it hasn't happened again in my test set up, but I haven't left the live exim 
running unattended.

So I don't think its my MX's DNS. It as if something has started going wrong 
with the resolver after a period of time.

But my second question is still open- why does Exim reject the local domain as 
unrouteable when this happens? What am I missing from my config to make it not 
do that?

James.
-- 
## List details at https://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