Spamassassin seems to still be working well. Once in a while (like once a day) I get a spam that slips through the net… then I go to the spam folder and see another 8 more that were caught. Nice.
Anyway, I’m saving the statistics below for my future reference.
After collecting 552 spams and 1053 not-spams, I erased the cache of learned messages, re-learned them all, then reprocessed them all as incoming. The spam was all at Bayes_70 and up and the good stuff was all at Bayes_30 and below (apart from one bounce from a spammer, bayes_44). Not surprisingly, after learning the target set, Bayes became quite good at processing the same set :)
So, the real test will be what happens with new mail coming in now. Speaking of which, right after reprocessing everything, there was one missed spam coming in at Bayes_50 with SA score 4.
----------------------------------------------------------------------
/home/gconnor/mail/Spam/track-spam
Bayes scores:
BAYES_00 0
BAYES_01 0
BAYES_10 0
BAYES_20 0
BAYES_30 0
BAYES_40 0
BAYES_44 0
BAYES_50 1
BAYES_56 0
BAYES_60 0
BAYES_70 10
BAYES_80 48
BAYES_90 493
BAYES_99 0
BAYES_.. 552
Spamassassin scores:
1 X-Spam-Level: *****
1 X-Spam-Level: ******
7 X-Spam-Level: *******
21 X-Spam-Level: ********
20 X-Spam-Level: *********
32 X-Spam-Level: **********
33 X-Spam-Level: ***********
44 X-Spam-Level: ************
38 X-Spam-Level: *************
46 X-Spam-Level: **************
36 X-Spam-Level: ***************
38 X-Spam-Level: ****************
31 X-Spam-Level: *****************
30 X-Spam-Level: ******************
28 X-Spam-Level: *******************
23 X-Spam-Level: ********************
26 X-Spam-Level: *********************
17 X-Spam-Level: **********************
12 X-Spam-Level: ***********************
9 X-Spam-Level: ************************
12 X-Spam-Level: *************************
4 X-Spam-Level: **************************
15 X-Spam-Level: ***************************
4 X-Spam-Level: ****************************
1 X-Spam-Level: *****************************
3 X-Spam-Level: ******************************
3 X-Spam-Level: *******************************
1 X-Spam-Level: ********************************
2 X-Spam-Level: *********************************
3 X-Spam-Level: **********************************
1 X-Spam-Level: ***********************************
2 X-Spam-Level: ************************************
2 X-Spam-Level: **************************************
1 X-Spam-Level: ****************************************
1 X-Spam-Level: *****************************************
1 X-Spam-Level: **********************************************
1 X-Spam-Level: ************************************************
1 X-Spam-Level: **************************************************
1 X-Spam-Level: ***********************************************************
552
----------------------------------------------------------------------
/home/gconnor/mail/Spam/missed-spam
Bayes scores:
BAYES_00 0
BAYES_01 0
BAYES_10 0
BAYES_20 0
BAYES_30 0
BAYES_40 0
BAYES_44 0
BAYES_50 1
BAYES_56 0
BAYES_60 0
BAYES_70 0
BAYES_80 0
BAYES_90 0
BAYES_99 0
BAYES_.. 1
Spamassassin scores:
1 X-Spam-Level: ****
1
----------------------------------------------------------------------
/home/gconnor/mail/Spam/track-not-spam
Bayes scores:
BAYES_00 143
BAYES_01 875
BAYES_10 29
BAYES_20 3
BAYES_30 2
BAYES_40 0
BAYES_44 1
BAYES_50 0
BAYES_56 0
BAYES_60 0
BAYES_70 0
BAYES_80 0
BAYES_90 0
BAYES_99 0
BAYES_.. 1053
Spamassassin scores:
1048 X-Spam-Level:
3 X-Spam-Level: *
1 X-Spam-Level: **
1 X-Spam-Level: ***
1053
----------------------------------------------------------------------
/home/gconnor/mail/Spam/missed-not-spam
Bayes scores:
BAYES_00 0
BAYES_01 0
BAYES_10 0
BAYES_20 0
BAYES_30 0
BAYES_40 0
BAYES_44 0
BAYES_50 0
BAYES_56 0
BAYES_60 0
BAYES_70 0
BAYES_80 0
BAYES_90 0
BAYES_99 0
BAYES_.. 0
Spamassassin scores:
0