Re: [AMaViS-user] is amavisd/milter supposed to forge a Received header for SpamAssassin?

2006-12-03 Thread Jo Rhett
Mark Martinec wrote: I'm still looking for some tests to confirm that all issues are fixed. I should add, as a general note: the fabricated Received header field should be the topmost Received in a file email.txt in a temporary directory (as prepared by a helper program). It should resemble

Re: [AMaViS-user] is amavisd/milter supposed to forge a Received header for SpamAssassin?

2006-12-01 Thread Jo Rhett
Petr Rehor wrote: There is a patch for testing (relative to amavisd-milter 1.2.0): http://home.i.cz/reho/amavisd-milter/received.diff release? -- Jo Rhett Network/Software Engineer Net Consonance - Take Surveys. Earn

Re: [AMaViS-user] is amavisd/milter supposed to forge a Received header for SpamAssassin?

2006-11-06 Thread Jo Rhett
Petr Rehor wrote: There is a patch for testing (relative to amavisd-milter 1.2.0): http://home.i.cz/reho/amavisd-milter/received.diff Any plans for a 1.2.1 release? -- Jo Rhett Network/Software Engineer Net Consonance -

Re: [AMaViS-user] is amavisd/milter supposed to forge a Received header for SpamAssassin?

2006-10-24 Thread Jo Rhett
At this point I'm going to say FIXED because the only things which are broken are part of SA's trusted networks breakage, and it requires some crafted forgery to hack through them. On Oct 17, 2006, at 5:06 PM, Jo Rhett wrote: On Oct 17, 2006, at 2:24 PM, Petr Rehor wrote: There is a patch

Re: [AMaViS-user] is amavisd/milter supposed to forge a Received header for SpamAssassin?

2006-10-18 Thread Mark Martinec
I'm still looking for some tests to confirm that all issues are fixed. I should add, as a general note: the fabricated Received header field should be the topmost Received in a file email.txt in a temporary directory (as prepared by a helper program). It should resemble the one that will later

Re: [AMaViS-user] is amavisd/milter supposed to forge a Received header for SpamAssassin?

2006-10-17 Thread Mark Martinec
No config option. A milter helper program (either the one supplied, or Petr Rehor's one) does insert a faked Received header by itself into a mail file it prepares for amavisd. Petr Rehor writes: Hmm. amavisd-milter doesn't do this now :-( I guess I was wrong. I thought it is based on the

Re: [AMaViS-user] is amavisd/milter supposed to forge a Received header for SpamAssassin?

2006-10-17 Thread Petr Rehor
On 10/17/06, Jo Rhett [EMAIL PROTECTED] wrote: Petr Rehor wrote: I'm going to investigate some solutions over the weekend. Which weekend? The weekend of 21-22? Ouch. I'll see if I can fabricate a hack patch sooner. There is a patch for testing (relative to amavisd-milter 1.2.0):

Re: [AMaViS-user] is amavisd/milter supposed to forge a Received header for SpamAssassin?

2006-10-17 Thread Jo Rhett
On Oct 17, 2006, at 2:24 PM, Petr Rehor wrote: There is a patch for testing (relative to amavisd-milter 1.2.0): http://home.i.cz/reho/amavisd-milter/received.diff I can confirm that it doesn't break anything. I'm still looking for some tests to confirm that all issues are fixed. -- Jo Rhett

Re: [AMaViS-user] is amavisd/milter supposed to forge a Received header for SpamAssassin?

2006-10-17 Thread Mark Martinec
I'm still looking for some tests to confirm that all issues are fixed. With 'amavisd debug-sa' the SA should report its analysis of Received header fields. Without Received from the MTA (or in case of a milter, a synthesized Received, because the actual one hasn't yet been generated), SA would

[AMaViS-user] is amavisd/milter supposed to forge a Received header for SpamAssassin?

2006-10-16 Thread Jo Rhett
So I've been tracking down some issues with SpamAssassin rulesets firing and/or misfiring, and we might have stumbled on the answer. SpamAssassin is apparently expecting that Amavis/Milter would forge a Received: header for the current connection. Does this jive with your understanding of

Re: [AMaViS-user] is amavisd/milter supposed to forge a Received header for SpamAssassin?

2006-10-16 Thread Mark Martinec
Jo, So I've been tracking down some issues with SpamAssassin rulesets firing and/or misfiring, and we might have stumbled on the answer. SpamAssassin is apparently expecting that Amavis/Milter would forge a Received: header for the current connection. Does this jive with your understanding