You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by Mike Burger <mb...@bubbanfriends.org> on 2004/07/14 20:37:54 UTC

Want to save hundreds on software? bdwc (fwd)

I'm not sure how this happened.

As can be seen by the report in the headers, this hits a whole lot of 
tests, but the score comes up as "nan" and ^X-SPAM-STATUS comes up as "no"

This is running SA 2.63.  Any ideas on how this scoring happened?

-- 
Mike Burger
http://www.bubbanfriends.org

Visit the Dog Pound II BBS
telnet://dogpound2.citadel.org or http://dogpound2.citadel.org:2000

To be notified of updates to the web site, visit 
http://www.bubbanfriends.org/mailman/listinfo/site-update, or send a 
message to:

site-update-request@bubbanfriends.org

with a message of: 

subscribe

Re: Want to save hundreds on software? bdwc (fwd)

Posted by David B Funk <db...@engineering.uiowa.edu>.
On Wed, 14 Jul 2004, Mike Burger wrote:

> I'm not sure how this happened.
>
> As can be seen by the report in the headers, this hits a whole lot of
> tests, but the score comes up as "nan" and ^X-SPAM-STATUS comes up as "no"
>
> This is running SA 2.63.  Any ideas on how this scoring happened?

Yes, software bug or hardware glitch.

There is a special internal format used to store floatingpoint numbers
in binary. Unlike integers, not all possible binary bit combinations
represent valid floatingpoint numbers. The IEEE floatingpoint spec takes
some of these invalid bit combinations and uses them as 'flags' to
represent special cases. NAN is one of these, it means "Not A Number".
This is used as the result of such operations as dividing by zero,
taking the square-root of a negative number, etc. Once a calculation
results in a NAN, further operations on that variable always result
in a NAN (IE it is defined to be a 'sticky' value).

The normal SA scoring calculations should not be capable of generating
a NAN, so my guess is that a software bug or hardware glitch somehow
resulted in an improper bit-pattern which was then interpreted as a NAN.

Take that message, run it thru SA again and see if you get the same
result (is that bug reproducable?)

-- 
Dave Funk                                  University of Iowa
<dbfunk (at) engineering.uiowa.edu>        College of Engineering
319/335-5751   FAX: 319/384-0549           1256 Seamans Center
Sys_admin/Postmaster/cell_admin            Iowa City, IA 52242-1527
#include <std_disclaimer.h>
Better is not better, 'standard' is better. B{