When I finally get round to installing Zimbra R5, I'll be moving a sub-domain over that uses a catch-all address and individual address blocking as a UCE prevention measure - I give each company/website an address <something>@sub.domain.tld and have all mail @sub.domain.tld go to the main incoming address @domain.tld, with a list of addresses that have started to attract spam set to fail quietly (not wasting bandwidth with bounce messages).

I've found ManagingDomains - Zimbra :: Wiki and How to create a /dev/null account or spam trap? - Zimbra :: Wiki in the wiki which will work for what I'm wanting, but I have two questions before I go ahead:

  1. Is there a more efficient way of blocking the bad addresses? I'm guessing the filter method will result in them going through normal spam and virus checks before being filtered and binned.
  2. The domain management page says that a catchall "can lead to being blacklisted". How it this so? I've not had any trouble (that I know of) with this sort of thing before, is there some new anti-spam/relay check mail providers are performing that a catch-all gives a false/bad readings for?