Re: Is p0f (passive OS detection) still usefull?

2014-07-04 Thread RW
On Fri, 04 Jul 2014 14:26:43 +0200 Christian Laußat wrote: > There is an old p0f plugin on > http://whatever.truls.org/spamassassin.text.shtml. > p0f didn't seem maintained since 2006, but there is a new rewritten > version on http://lcamtuf.coredump.cx/p0f3/ (with a different API). > > I have

Re: Pyzor with aliases.

2014-07-04 Thread Steve Bergman
On 07/04/2014 11:21 AM, Matus UHLAR - fantomas wrote: there is no reason why sa-milter should know about users - it would require adding new useless code to it. Especially not when there is no local user, which is something spamd must take care about, so there's no need to duplicate this job

Re: Pyzor with aliases.

2014-07-04 Thread Matus UHLAR - fantomas
On 07/04/2014 07:02 AM, Matus UHLAR - fantomas wrote: I have read the man page multiple times. spamass-milter passes the local part of mail recipient to spamd, which is not necessarily a username. spamass-milter even can call "sendmail -bv" to get the user, but it still may not be local user.

Re: Pyzor with aliases.

2014-07-04 Thread Steve Bergman
On 07/04/2014 07:02 AM, Matus UHLAR - fantomas wrote: I have read the man page multiple times. spamass-milter passes the local part of mail recipient to spamd, which is not necessarily a username. spamass-milter even can call "sendmail -bv" to get the user, but it still may not be local user.

Re: Is p0f (passive OS detection) still usefull?

2014-07-04 Thread David F. Skoll
On Fri, 04 Jul 2014 14:26:43 +0200 Christian Lau__at wrote: > Do you think passive OS detection is still usefull today? It's mildly useful. We throw p0f results into Bayes, and indeed, "Windows XP" is a pretty strong spam indicator. But it's not a magic bullet. Regards, David.

Re: Is p0f (passive OS detection) still usefull?

2014-07-04 Thread John Hardin
On Fri, 4 Jul 2014, Christian Laußat wrote: There is an old p0f plugin on http://whatever.truls.org/spamassassin.text.shtml. p0f didn't seem maintained since 2006, but there is a new rewritten version on http://lcamtuf.coredump.cx/p0f3/ (with a different API). I have written a plugin to use t

Is p0f (passive OS detection) still usefull?

2014-07-04 Thread Christian Laußat
There is an old p0f plugin on http://whatever.truls.org/spamassassin.text.shtml. p0f didn't seem maintained since 2006, but there is a new rewritten version on http://lcamtuf.coredump.cx/p0f3/ (with a different API). I have written a plugin to use this new version of p0f ( https://kvm.laussat.

Re: Pyzor with aliases.

2014-07-04 Thread Matus UHLAR - fantomas
On 07/04/2014 05:27 AM, Matus UHLAR - fantomas wrote: I repeat: spamass-milter does not (and can not) know about local users. On 04.07.14 06:30, Steve Bergman wrote: SPAMASS_MILTER(8) I have read the man page multiple times. spamass-milter passes the local part of mail recipient to spamd, wh

Re: Pyzor with aliases.

2014-07-04 Thread Steve Bergman
On 07/04/2014 05:27 AM, Matus UHLAR - fantomas wrote: > I repeat: spamass-milter does not (and can not) know about local users. SPAMASS_MILTER(8) -u defaultuser Pass the username part of the first recipient to spamc with the -u flag. This allows user preferences files to be used. If the mess

Re: Pyzor with aliases.

2014-07-04 Thread Matus UHLAR - fantomas
https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6995 On 07/03/2014 09:38 AM, Matus UHLAR - fantomas wrote: I see you are next person (after me) who encountered this issue. Maybe you could comment (or at least +1) the bug. On 03.07.14 09:57, Steve Bergman wrote: What do you think woul