On 01/20/2014 12:43 PM, Matus UHLAR - fantomas wrote:
On 01/20/2014 11:37 AM, Matus UHLAR - fantomas wrote:
I heave received a ham message, that hit BAYES_50 at the time I
received
it. After feeding to "spamassassin -k" and rechecking it gives
BAYES_99: Unfortunately, feeding message to sa-learn explains it
was not
learntL
OK, an update: --forget and --ham do show that tokens were forgotten /
learned, but no change in result (score: 1.000)...
On 20.01.14 12:02, Axb wrote:
this probably means that the msg hits too many spammy tokens so that
"learning one" is not making the big difference.
I wonder why did the message have BAYES_50 just a few hours ago when it was
checked forthe first time....
Are you doing regular expiration runs?
trying force-expire changed nothing...
Try adding token reports to your scanned msgs.
(http://spamassassin.apache.org/full/3.3.x/doc/Mail_SpamAssassin_Conf.txt)
_HAMMYTOKENS(N)_ the N most significant hammy tokens (default, 5)
_SPAMMYTOKENS(N)_ the N most significant spammy tokens (default, 5)
report Report _SPAMMYTOKENS(25)_
report Report _HAMMYTOKENS(25)_
That may help you trouble shoot cases like this
