Hi,
First of all - thanks so much for the help! Much appreciated. Some thoughts, below. 1) I can ask at spamassassin, but it seems like my spamassassin settings aren't being used. Does amavis somehow take this over, or should the spamassassin settings really work? Sounds like a dumb question I know, but I have tried changing the settings and they don't seem to be used (and I'm restarting spamassassin and amavis both ... :-(). 2) You are correct - it's because my IP is a dynamic IP, provided by Verizion FIOS. They have added all their IP's to the blacklist. I just have a backup mail server (at my brother's house), that forwards email to me when my main server goes down. But I don't want the email blocked just because it came from FIOS (in fact, one of the failing RBL checks just says it's a Verizon IP, which I already know ... :-)). 3) Sorry, just trying to understand your last comment (about internal and trusted networks). Will the previous relays be checked, just the last one is skipped? And can I use a machine name (FQDN), or do I have to enter the IP address (it's dynamic, so name would be nice). I will check the reference you mention, just not sure my spamassassin settings are being used ... :-( Thanks again! ... Russell On 2012-12-30 02:42, Cedric Knight wrote: > Hi Russell > > On 30/12/12 05:21, R. Morris wrote: > >> Hi, This may be a dumb question (and likely is, sorry!), but I haven't been able to figure out how to get it working, so ... > > Well, I hope this isn't a dumb answer. It does sound like a > SpamAssassin issue maybe best dealt with at [email protected]. > >> I have a working setup with Postfix + Amavis-new. Works quite well, except for one thing - if email goes through my backup MX (mail server), then it is flagged as spam (due to RBL checks). I can disable all checking based on this IP, but I really only want to disable RBL / DNSBL checks for this IP, and let the other checks proceed as usual. > > Which RBL checks, and why is your backup MX on an RBL? Maybe it's just > a list of dynamic or consumer IPs, but maybe it's associated with a > previous exploit.
