Karsten Bräckelmann a écrit : > On Tue, 2009-01-20 at 15:29 +0000, Arthur Dent wrote: >> On Tue, Jan 20, 2009 at 03:48:55PM +0100, Karsten Bräckelmann wrote: > >>> I explained it slightly more detailed in Bug 6008. >>> https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6008 >> Ahhh.. I see (er I think...) >> >> Tell me; when you reported that bugzilla was hitting ANY_BOUNCE_MESSAGE did >> it >> hit ANY_BOUNCE_MESSAGE? ;) ...er... OK I'll shut up now... > > No. I didn't even notice that myself, but have been asked about it.
I did and it hit. I should have said so on the bug page... I no more use vbounce rules (well, I enable them just to see the results in headers). a long time ago, I posted a case where this hits a message that is not a bounce at all (the body discussed a bounce). I didn't find a simple way to differentiate betwenn such "bounce discussions" and real bounces (unfortunately, many MTAs use non-standard formats, if they use a format at all!). > That's because I do *not* scan bugzilla mail for spam, but filter it out > early. > > There's absolutely no reason to torture my SA with the bulk of bugzilla > or mailing-list mail. Yes, I do a *lot* of both. They get delivered > without being scanned by SA. And I highly recommend doing this. > > However, yes, of course, it *would* have hit ANY_BOUNCE_MESSAGE... ;) > > guenther >
