On Mon, 2 Apr 2007, J. wrote:
> > I don't know if there's a consensus or not among listmembers, but we
> > regularly see someone wondering why things are all clogged up when
> > autoexpire reaches the point that it takes longer to expire old
> > tokens than the process that's trying to filter m
--- "John D. Hardin" <[EMAIL PROTECTED]> wrote:
> On Mon, 2 Apr 2007, Dan Barker wrote:
>
> > Autoexpire will keep the db size approximately constant, even when
> > your totals get higher. Otherwise, it would grow without bound.
> > Disk is cheap, but not free!
>
> Yabbut time can become *very*
On Mon, 2 Apr 2007, Dan Barker wrote:
> Autoexpire will keep the db size approximately constant, even when
> your totals get higher. Otherwise, it would grow without bound.
> Disk is cheap, but not free!
Yabbut time can become *very* expensive.
I don't know if there's a consensus or not among li
@spamassassin.apache.org
Subject: RE: YA sa-learn question
--- Dan Barker <[EMAIL PROTECTED]> wrote:
> Yes, dump magic and note that the nspam and nham increase.
>
> -Original Message-----
> Subject: YA sa-learn question
>
> I thought I had solved my sa-learn issues by always
--- Dan Barker <[EMAIL PROTECTED]> wrote:
> Yes, dump magic and note that the nspam and nham increase.
>
> -Original Message-----
> Subject: YA sa-learn question
>
> I thought I had solved my sa-learn issues by always doing su qscand
Thanks, that worked. Does thi
iry atime
0.000 0 86400 0 non-token data: last expire atime delta
0.000 0 56486 0 non-token data: last expire reduction count
-Original Message-
From: J. [mailto:[EMAIL PROTECTED]
Sent: Sunday, April 01, 2007 1:47 PM
To: users@spamassassin.apache.org
Subject: YA sa-le
I thought I had solved my sa-learn issues by always doing su qscand
whenever I want to run sa in debug mode or run sa-learn. Today I got
some false positives and some false negatives and decided to train the
bayes db using them. After I ran sa-learn for both ham and spam I
looked at the bayes db's
