Theo wrote:
> In 2.5x, if you're going to do manual expires, change the expiry_count
> value to something really large.
>
> In 2.6x, just do 'bayes_auto_expire 0'. ;)
Okay, so I just set the bayes_expiry_scan_count to 500000 (as Dallas
suggested in his previous email, and you do here). No expirations are done
automatically, but then according to the sa-learn man page:
SpamAssassin runs through every token in the database. If that token
has
not been used during the scanning of the last
`bayes_expiry_scan_count'
messages, it is marked for deletion.
So if I have 500,000 for bayes_expiry_scan_count, then a token will only be
marked for deletion very rarely, as it's unlikely that most of the tokens
will *not* have been seen in the last 1/2 million emails.
Next, if that operation would bring the number of tokens below the
`bayes_expiry_min_db_size' threshold, it removes tokens from the for-
deletion list until the resulting database would contain
`bayes_expiry_min_db_size' token entries.
It then removes the listed tokens and updates the 'last expiry'
setting.
So this explains what happens if the DB gets too *small*. But what governs
how *big* the database can get? Would it just be big enough to contain the
last 500,000 seen tokens? Would this be a problem for performance/DB size?
thanks!!
johnS
-------------------------------------------------------
This SF.net email is sponsored by: eBay
Get office equipment for less on eBay!
http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
