On Sat, 22 Dec 2007, Phil Pennock wrote:
> So you can rotate all you like; provided that you retry from the same IP
> at least once, more than 10 minutes after the first attempt from that
> IP, then it will go through.

I just had this section in each of the server;s configration, of course
the fallback host is the next in the line.
Of course this is useful if one server cannot connect, usally for network
congestion, to a determinate remoe server, and rathere than waith and
retry would immediately pass the burden to another (physical or virtual)
server that uses a different routing (and so a different IP).
In this case the arrangement speed thing up, also because the last server
in the chain is an olt machine whose purpose is only to retry (but only
after the third try).
In the case of a 4xx error the rule is "try later, will be more luky", not
the dreaded "Please hold to maintain your priority" neither "try later,
but from the same IP".
Am i wrong ?
what is the ratio between messages delayed due a network/server overload
(in wich case my way usually speed up things) and message delayed due to
greylisting (that anyway would create an "istititional" delay) ??


lookuphost:
  driver = dnslookup
  domains = ! +local_domains
  pass_on_timeout
  self = fail
  fallback_hosts = mx3.domai.in
  transport = remote_smtp
  no_more


> 
> -- 
> ## List details at http://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/
> 
> 


-- 
## List details at http://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