Matt wrote: > Go figure, I rebooted the relay server, and now the queues are flushing > like crazy, almost down to 2000 now (they got up to 4000 this afternoon).
> The server load is 11.00+ and the memory and CPU is maxed out right now > as its processing the mail. Maybe it just needed some kind of jump > start? Not sure why.... Memory maxed out is never a good thing. I really like to run enough memory that I rarely use swap space at all. I do think adding more memory would help you. The server may have been unresponsive due to excessive use of swap space, and rebooting cleaned it up for a short time. > Also, relay1 is the SAME machine as relay2 and relay3, yet it always > seems backed up. Is this because its the first server listed? Should I > try rotating relay1 to the end of our pool for some domains so that, if > spammers and/or non-compliant mail servers pull the first MX record > instead of rotating, could this mean something? This is puzzling. I would think there would be no difference in the load on the servers, but this does depend on what the name servers serve up. It may be a good idea to manually provide different priorities on some of your domains (make relay1 the highest, or even leave it out). How are you determining the number of messages received by each server? Maybe they are receiving the same number now, but relay1 cannot handle them as well as the others because of some as yet unseen problem. I would spend some time reading your maillog for clues to problems. > Thanks! > -Matt Gary V ------------------------------------------------------- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl _______________________________________________ AMaViS-user mailing list AMaViS-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/amavis-user AMaViS-FAQ:http://www.amavis.org/amavis-faq.php3 AMaViS-HowTos:http://www.amavis.org/howto/
