Reason for rejected message ?

Vernon Schryver vjs@calcite.rhyolite.com
Thu Mar 1 21:04:41 UTC 2007


> > An oddity I saw while installing SUSE and then trying to debug these
> > problems was that the X server ate up 97.2% of a CPU for hours on end.
>
> That leaves 2.8% to dcc and other tasks. That's plenty on a modern system...

for some value of "plenty"...I did some little grumbling about what
SUSE had done to make gcc so much slower compiliing the DCC source
before I noticed the X craziness.


Are there any more clues about the original problem of spurious dccifd
rejections?  I looked at a lot of code and shot down a lot of theories
about how it might happen, including my first theory that quote problem
in dcc_conf was making something look like a spam threshold of 0.

How often does it happen?  
Is there anything like a reproducable test case?
I've assumed that it happens even when the Fuz2 checksum is computed. 
    Is that true?
Does it happen when the SMTP client IP address is not marked "mx"?
   (There is a separate bug there that I have reproduced and think I
   understand.  You should never see "mx" on the "IP:" line except when
   there are Received: lines in the message body that make sense.  That
   leads me to ask "Are the Yahoo Received lines standard or do I need
   to add yet more complications to the Received line parsing?)


Vernon Schryver    vjs@rhyolite.com



More information about the DCC mailing list

Contact vjs@rhyolite.com by mail or use the form.