dean at iglou
Feb 24, 2009, 8:34 AM
Post #2 of 2
(738 views)
Permalink
|
------- You are receiving this mail because: -------
You are on the CC list for the bug.
http://bugs.exim.org/show_bug.cgi?id=815
Dean Brooks <dean [at] iglou> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |dean [at] iglou
--- Comment #2 from Dean Brooks <dean [at] iglou> 2009-02-24 16:34:28 ---
Fallback hosts is a feature of transports if they are unable to successfully
deliver the message AFTER it has been routed. The fallback_hosts on routers
is, per the documentation, just a way to pass along a value for the
fallback_hosts transport option.
What you are talking about is a routing failure, which is not what the
fallback_hosts option is for. It could be changed, but it would not be
trivial.
Instead, there is a much simpler way to do this:
process_and_forward:
driver = manualroute
domains = +preprocess_domains
transport = process_and_forward_smtp
route_list = *
${lookup{$domain}dbm{/etc/exim/control/run/preprocess.db}{$value}{}}
fallback_hosts = retry.ctyme.com
pass_on_timeout
no_more
send_to_fallback:
driver = manualroute
domains = +preprocess_domains
route_data = retry.ctyme.com
transport = process_and_forward_smtp
Transport failures (i.e. 4xx response codes from remote host) will continue to
work using your original router by using the fallback_hosts option.
However, I added the "pass_on_timeout" option to your first router. If that
router gets a DNS timeout, it will pass on to the next router. The next router
is a catch-all, so to speak, that will ONLY catch DNS timeouts due to the
no_more option on the first router. It simply forces the email on to the
fallback host.
I too wish that fallback_hosts worked the way you want it to, but its not a
bug, that is simply a feature request.
--
Configure bugmail: http://bugs.exim.org/userprefs.cgi?tab=email
--
## List details at http://lists.exim.org/mailman/listinfo/exim-dev Exim details at http://www.exim.org/ ##
|
