Rule 'TVD_FROM_1' causing all our emails' spam scores go up

Hi SpamAssassin PMC, My name is Henry Draper. I'm from Comm100 Network Corporation. We use SpamAssassin to detect our emails' spam scores. l how you're invoking SpamAssassin - with default configurations l what version of SpamAssassin you're using - Version 3.3.2 l your platform. - Windows

Re: MariaDB instead of MySQL

Yeah - I'm running global too. Not per user. On 5/16/2013 5:38 PM, Rick Macdougall wrote: Hi, One common bayes DB running MySql on one server. Spa001 running SA and MySQL plus auto learn Spa010 running SA no auto learn Spa011 running SA no auto learn Spa013 running SA with auto learn All Spas

Re: MariaDB instead of MySQL

On 5/16/2013 5:02 PM, Alex wrote: Hi, Running a 4 server cluster of SA servers with MySQL running on one. 2 running auto learn and the other two just doing queries. Looks like I average about 6300 queries a minute. No noticeable problems with it. So they are all separate from each other,

Re: MariaDB instead of MySQL

Hi, One common bayes DB running MySql on one server. Spa001 running SA and MySQL plus auto learn Spa010 running SA no auto learn Spa011 running SA no auto learn Spa013 running SA with auto learn All Spas looking at the MySQL server on Spa001. Global bayes database btw, not per user as we are an

Re: MariaDB instead of MySQL

Hi, > Running a 4 server cluster of SA servers with MySQL running on one. 2 > running auto learn and the other two just > doing queries. Looks like I average about 6300 queries a minute. No > noticeable problems with it. So they are all separate from each other, not sharing a common bayes db

Re: MariaDB instead of MySQL

Hi, Running a 4 server cluster of SA servers with MySQL running on one. 2 running auto learn and the other two just doing queries. Looks like I average about 6300 queries a minute. No noticeable problems with it. Recently looked at Maria but we are in the middle of a move here, would love to

MariaDB instead of MySQL

In the past I have never managed to get bayes to work with MySQL in a reliable way. Even running in ram it would die of some sort of corruption after a few hours. Now I'm trying the MariaDB replacement. I will admit I'm running a faster server than the last time I tried MySQL - but so far so g

Re: N00b question - sa-milter Checking Orig Networks

On Wed, 15 May 2013 08:24:24 +0200 Matus UHLAR - fantomas wrote: > On 14.05.13 20:55, Martin Gregorie wrote: > >As to the letting SA check the origination addresses, my setup looks > >like this: > > > >internal_networks192.168.7/24 > > > >trusted_networks 192.168.7/24 > >trusted_networks

Re: Bayes & sa-learn training question - How to exclude certain URLs found in emails

On 16.05.13 19:32, Simon Loewenthal wrote: I shall have Bayes forget emails, then retrain with a rule that counts linkedin invitations as spam. I have one I wrote earlier and implemented after Bayes had gone off track. You don't have to forget mail, unless you have feed incorrectly formatted m

Re: Bayes & sa-learn training question - How to exclude certain URLs found in emails

Thank-you everyone. I shall have Bayes forget emails, then retrain with a rule that counts linkedin invitations as spam. I have one I wrote earlier and implemented after Bayes had gone off track. Cheers, S. -- as silly as fun simon@klunky / .co.uk / .org / .net pgp 4BA78604 Benny Pedersen

Re: Bayes & sa-learn training question - How to exclude certain URLs found in emails

On Thu, 16 May 2013 18:07:53 +0200 Benny Pedersen wrote: > Simon Loewenthal skrev den 2013-05-16 15:17: > > > * have sa-learn exclude references to linkedin? > > basicly you need to know how bayes works, it does not just use the > word linkedin as writed, but more or less split it to each letter

Re: Bayes & sa-learn training question - How to exclude certain URLs found in emails

Simon Loewenthal skrev den 2013-05-16 15:17: * have sa-learn exclude references to linkedin? basicly you need to know how bayes works, it does not just use the word linkedin as writed, but more or less split it to each letter, so linkedin and other content is splited into atoms that being st

Re: Bayes & sa-learn training question - How to exclude certain URLs found in emails

On 16.05.13 15:17, Simon Loewenthal wrote: I turned shortcircuit for BAYES_00 on a server, and noticed that LinkedIn invitation emails hit BAYES_00. A bit strange I thought being unlikely someone had run sa-learn on LinkedIn emails. I grepped on all the Ham and Spam directories and hit lots of

Re: Bayes & sa-learn training question - How to exclude certain URLs found in emails

On 2013-05-16 16:17, RW wrote: On Thu, 16 May 2013 15:17:14 +0200 Simon Loewenthal wrote: Hi all, I turned shortcircuit for BAYES_00 on a server, and noticed that LinkedIn invitation emails hit BAYES_00. When you say email, I presume you mean spam. Yep A bit strange I thought being unl

Re: Bayes & sa-learn training question - How to exclude certain URLs found in emails

On Thu, 16 May 2013 15:17:14 +0200 Simon Loewenthal wrote: > Hi all, > > I turned shortcircuit for BAYES_00 on a server, and noticed that > LinkedIn > invitation emails hit BAYES_00. When you say email, I presume you mean spam. > A bit strange I thought being > unlikely someone had run sa-

Bayes & sa-learn training question - How to exclude certain URLs found in emails

Hi all, I turned shortcircuit for BAYES_00 on a server, and noticed that LinkedIn invitation emails hit BAYES_00. A bit strange I thought being unlikely someone had run sa-learn on LinkedIn emails. I grepped on all the Ham and Spam directories and hit lots of "linkedin,com" in URLs in the bod