Why don´t revert to the old behavior ? And enable BAYES_999 only for SA
3.4.1 +
I run SA 3.3.2 on a few servers and I don´t want patch every one.
El 17/02/2014 21:04, Kevin A. McGrail escribió:
On 2/17/2014 8:34 PM, Ian Zimmerman wrote:
On Mon, 17 Feb 2014 20:21:47 -0500
"Kevin A. McGrail"
2014-02-18 02:34, Ian Zimmerman wrote:
On Mon, 17 Feb 2014 20:21:47 -0500
"Kevin A. McGrail" wrote:
Kevin> https://issues.apache.org/SpamAssassin/show_bug.cgi?id=7013 is
Kevin> open on the issue. Should be a very small patch if you want to
Kevin> update your installation.
So, this means that
On 2/17/2014 8:34 PM, Ian Zimmerman wrote:
On Mon, 17 Feb 2014 20:21:47 -0500
"Kevin A. McGrail" wrote:
Kevin> https://issues.apache.org/SpamAssassin/show_bug.cgi?id=7013 is
Kevin> open on the issue. Should be a very small patch if you want to
Kevin> update your installation.
So, this means t
On Mon, 17 Feb 2014 20:21:47 -0500
"Kevin A. McGrail" wrote:
Kevin> https://issues.apache.org/SpamAssassin/show_bug.cgi?id=7013 is
Kevin> open on the issue. Should be a very small patch if you want to
Kevin> update your installation.
So, this means that the BAYES_999 mixup is not just a tempora
Good catch.
https://issues.apache.org/SpamAssassin/show_bug.cgi?id=7013 is open on
the issue. Should be a very small patch if you want to update your
installation.
Regards,
KAM
On 2/17/2014 6:45 PM, Alex R wrote:
Look like BAYES_999 rules ignore the bayes_auto_learn_on_error 1 setting
And
On 2/17/2014 8:12 PM, Mark Martinec wrote:
2014-02-18 Alex R wrote:
Look like BAYES_999 rules ignore the bayes_auto_learn_on_error 1 setting
And keep learning spam.
--- lib/Mail/SpamAssassin/Plugin/AutoLearnThreshold.pm (revision
1569155)
+++ lib/Mail/SpamAssassin/Plugin/AutoLearnThreshold.
2014-02-18 Alex R wrote:
Look like BAYES_999 rules ignore the bayes_auto_learn_on_error 1
setting
And keep learning spam.
--- lib/Mail/SpamAssassin/Plugin/AutoLearnThreshold.pm (revision
1569155)
+++ lib/Mail/SpamAssassin/Plugin/AutoLearnThreshold.pm (working copy)
@@ -241,7 +241,8 @@
Look like BAYES_999 rules ignore the bayes_auto_learn_on_error 1 setting
And keep learning spam.
example
spamd: result: Y 34 -
AWL,AXB_XMAILER_MIMEOLE_OL_024C2,BAYES_999,DKIM_ADSP_CUSTOM_MED,FORGED_MUA_OUTLOOK,FORGED_OUTLOOK_HTML,FREEMAIL_ENVFROM_END_DIGIT,FREEMAIL_FROM,FREEMAIL_REPLYTO,FROM_