On Tue, August 12, 2008 02:48, Michael Scheidell wrote: > I was looking at the sql commands used to expire bayes data and noticed > some indexes that combined 'id' with token and atime. > (two indexes, on bayes_token)
bayes can expire it self, no need for help from amavisd-new :) > and noticed id also on bayes_vars. > > i tried to see what the 'id' value was used for, since every row in both > tables had id=1. > select count(*) from bayes_token where id !=1; > and > select count(*) from bayes_vars where id != 1; > > Both come back '0'? when amavisd-new uses spamassassin its just single user bayes so no need to report more then one id, but that say spamassassin can still work multiuser in sql if bayes is not setup to use single user perldoc Mail::SpamAssassin::Plugin::Bayes perldoc Mail::SpamAssassin::Conf > is this for multi-user bayes ? anyone using amavisd-new know if there > is any real reason not to eliminate the id based keys and reindex on > atime and token seperatly? its not practical to do this tune up since we dont want a spamassassin only works with amavisd-new > wouldn't that make sa-learn --force-expire faster? make writes (updates) > to bayes table faster with one less combination key? (two less on > bayes_token)? yes it will, but it will be on cost of flexibilities -- Benny Pedersen Need more webspace ? http://www.servage.net/?coupon=cust37098 ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ AMaViS-user mailing list AMaViS-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/amavis-user AMaViS-FAQ:http://www.amavis.org/amavis-faq.php3 AMaViS-HowTos:http://www.amavis.org/howto/
