I'm intruiged by no SQL being required Fritz! If you have some time, some explanation would be great:
1) Am I correct that the 2 slaves simply copy the files folder, spamdb, spamdb.helo, and the white/red list from the master? 2) What happens when server 2 or 3 receive a message from a whitelisted address that should be renewing the TTL for that whitelist entry? How does the master find out about this? 3) What happens when a whitelisted address sends something through server 2 or 3 that has a new address as a cc that should be whitelisted (since it's come from a white address)? 4) If servers 2 and 3 don't store any mail, how do messages that come in through them contribute to the corpus? 5) What do your mx records look like? Thanks. This could be very helpful for a mid sized non-profit group that I'm trying to get to use ASSP. On Thu, Apr 22, 2010 at 12:23 PM, Fritz Borgstedt <[email protected]> wrote: > Actually I have a very similar set up running in ASSP V1 with 3 > servers for incoming and 1 for incoming/outgoing. > It works great and I do not see any limit to the workload it can > handle. > > The ASSP handling the incoming/outgoing mail is the "master", the 3 > incoming are working as "slaves". > No SQL involved. > > The master does: > -email-interface > -block reporting > -rebuildspamdb (every hour) > -whitelist/redlist save every 10 minutes > > The slaves do: > sync > - files folder > - whitelist/redlist (if not in files folder) > - spamdb > - spamdb.helo > The slaves do not: > - store any spam/notspam themselves > > > ------------------------------------------------------------------------------ > _______________________________________________ > Assp-user mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/assp-user > ------------------------------------------------------------------------------ _______________________________________________ Assp-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/assp-user
