Re: Help eliminate false positive for Google Code notifications
Benny Pedersen wrote: its was good since to many still use it :) In my case it was that the old rulesets were left behind long after the updates stopped; they kept getting transferred over through upgrades of SpamAssassin and Perl. Once I deleted them, all was well. Well, except that more spam started getting through. :)
Re: Help eliminate false positive for Google Code notifications
Mike Brown skrev den 2013-07-11 18:09: SARE_MSGID_DDDASH Message-ID has ratware pattern (9-, 9$, 99-) sare rulesets is depricated, so you are on your own :=)
Re: Help eliminate false positive for Google Code notifications
Axb skrev den 2013-07-11 18:13: a retired SARE Ninja its was good since to many still use it :)
Help eliminate false positive for Google Code notifications
Google Code sends out notifications from project name@googlecode.com. These notifications have Message-ID headers that start with two digits and a dash, triggering this rule: SARE_MSGID_DDDASH Message-ID has ratware pattern (9-, 9$, 99-) The rule was proposed in 2004: https://mail-archives.apache.org/mod_mbox/spamassassin-users/200402.mbox/%3c20040204190450.9b96217...@jmason.org%3E A sample Message-ID (I have an issue starred in the Android project): 46-1531741276455824-7215198307142895543-android=googlecode@googlecode.com Complete mbox message at http://pastebin.com/W5cN4DFd The false positive is not contributing much to the score (1.666), but I don't like it, so I'd like to avoid triggering the rule altogether if I can. I want to do it in the preferred way, if there is a preferred way. Any solution I would come up with would be pretty kludgy. So, suggestions appreciated! Thanks.
Re: Help eliminate false positive for Google Code notifications
On 07/11/2013 06:09 PM, Mike Brown wrote: Google Code sends out notifications from project name@googlecode.com. These notifications have Message-ID headers that start with two digits and a dash, triggering this rule: SARE_MSGID_DDDASH Message-ID has ratware pattern (9-, 9$, 99-) The rule was proposed in 2004: https://mail-archives.apache.org/mod_mbox/spamassassin-users/200402.mbox/%3c20040204190450.9b96217...@jmason.org%3E A sample Message-ID (I have an issue starred in the Android project): 46-1531741276455824-7215198307142895543-android=googlecode@googlecode.com Complete mbox message at http://pastebin.com/W5cN4DFd The false positive is not contributing much to the score (1.666), but I don't like it, so I'd like to avoid triggering the rule altogether if I can. I want to do it in the preferred way, if there is a preferred way. Any solution I would come up with would be pretty kludgy. So, suggestions appreciated! Thanks. SARE rules are obsolete/unsupported/ancient/history/etc and shouldn't be used. Do yourself a favour and remove those files - will save you CPU cycles, memory and lots of headaches. your truly, a retired SARE Ninja
Re: Help eliminate false positive for Google Code notifications
Axb wrote: SARE rules are obsolete/unsupported/ancient/history/etc and shouldn't be used. Do yourself a favour and remove those files - will save you CPU cycles, memory and lots of headaches. Heh, even easier than I thought. I think I had assumed that if I stopped fetching them, I wouldn't have them anymore, especially after upgrading Spamassassin. But they stayed and got copied over from upgrade to upgrade. Thanks!
