Author: Evgeniy Berdnikov Date: To: exim-users Subject: Re: [exim] DNS problems with sending via multiple smarthosts
On Wed, Jul 17, 2019 at 02:31:59PM +0930, David Purton via Exim-users wrote: > On Tue, Jul 16, 2019 at 02:43:12PM +0100, Jeremy Harris via Exim-users wrote: ... > > ( and grab stderr to a file; it'll be long.
> > "exim -d+all -qf 2>&1 | tee log" )
> Thanks. The relevant part of the resulting log is for a similar message
> and the same network is: ...
> 13:34:13 2071 set transport remote_smtp_smarthost
> 13:34:13 2071 finding IP address for smtp.gmail.com:587
> 13:34:13 2071 host=smtp.gmail.com port=587
> 13:34:13 2071 calling host_find_byname
> 13:34:13 2071 Coerced resolver DNSSEC support on.
> 13:34:24 2071 gethostbyname2(af=inet6) returned 1 (HOST_NOT_FOUND)
> 13:34:35 2071 gethostbyname2(af=inet) returned 1 (HOST_NOT_FOUND)
> 13:34:35 2071 no IP address found for host smtp.gmail.com
I propose to install tcpdump and run it in parallel with "exim -qf":
tcpdump -nlUv -s0 -i any port domain
Then look what happens on network level while exim tries to resolve
"smtp.gmail.com". I suspect that there are some local resolver suffixes
(from WiFi provider) in /etc/resolv.conf, and they lead to attempts
to resolve non-existent domains like "smtp.gmail.com.local.tld".
Suffixes are tried first by resolver (before attempts to resolve name
as it is), so they may lead to NXdomain errors.
This message was posted to the following mailing lists: