Jason Frisvold wrote: > On 1/12/06, Matthias Keller <[EMAIL PROTECTED]> wrote: >> Sorry i accidently sent the previous (incomplete) message... >> >> I'm by no means a bayes specialist but i dont think it's a good idea >> just to delete the oldest entries since SA provides its own mean of >> purging... > > Gotcha.. I kinda knew about it, and thought I had it enabled (I do) > .. Any recommendations on the number of tokens to keep? > > I think I've determined part of the problem. Non-existant users are > still in the database! ACK! > > Given that, is it safe to walk through the database and delete all > data for non-existant users? It looks like the tables are linkes via > a field called id... >
You shouldn't check mail for non-existent users. The best way to clear this data is via sa-learn. sa-learn -u <nonexistent user> --clear You can get a list of usernames from the bayes_vars table, it is not the id column it is the username column. If you don't really have the ability to not check mail for non-existent users you might want to check out this plugin: http://wiki.apache.org/spamassassin/AuthzUserPlugin Michael
signature.asc
Description: OpenPGP digital signature
