Hello all, I am about to discuss a antispam/antivirus policy with a client of mine. There are however a couple of "gray areas", for me, in the documentations. I will be Using an ldap server where users reside and posibly a web interface for user preferences. The questions that have risen are:
1. From the preferences stated here: http://spamassassin.apache.org/full/3.0.x/dist/doc/Mail_SpamAssassin_Conf.html#user_preferences (and also the rest of them) how many can I delegate to the ldap backend? are there any spamassassin specific limitations? Does the Ldap subsystem use any kind of caching mechanism or works in a one message, one recipient, one query way, has anyone noted what happens in multiple recipients per message? 2. The final one is a bit of performance oriented question the next part of the equation is the milter that will be used. I would like to hear some opinions regarding the way they operate and specificly with multiple rcpt to headers used? Thanks again for the time ... -- Vaggelis Georgatos IT&T Specialist
