On Wed, 15 Feb 2012, Tim Connors <tconnors@rather.puzzling.org> wrote:
And make it impossible to implement certain types of highly available MTA clusters. What happens when the MTA that sent out the email that you have greylisted, then crashes before it gets a chance to try again? Should no other MTA in the cluster be allowed to pull the message off the spool on shared storage and retry itself?
If the cluster was failing over often enough for recipients around the Internet to notice this then there would be some serious problem. I don't think that the results of different IP addresses causing problems with grey-listing are relates to cluster failover, but to using multiple addresses randomly when there's no need to do it. Even if one designed a system where email came from a database (the most likely way of having email totally disassociated from the IP address of the sending system) then one could have a NAT system in front of the cluster that made it have a single source IP to the rest of the world. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/