If we have eliminated that recent canonicalization bug (it's only present 
in the "relaxed" body canonicalization code, which Mark isn't using) and 
the spaces-in-key-records question (I agree with Mark, that's probably a 
false lead), I'd encourage you guys to use the debugging features in your 
respective filters to save the canonicalized forms and diff them.  You can 
either:

(a) both set the DKIMDEBUG environment variable to "c" and restart the 
filter, saving canonicalized forms to /var/tmp; then exchange and "diff" 
them to spot the differences; OR

(b) Mark can set DKIMDEBUG as above, Tony can restart his filter with 
"-R", and then Mark can set an "r=" tag in his SSP record to contain a 
debugging address, and then Mark can send a message to Tony; if it 
fails, Tony's filter will reply with its canonicalized forms in a separate 
message and Mark can diff them to what his side saved

(c) Mark can set "Diagnostics" to "True" in his configuration file (adding 
z= tags to the signatures he generates; these can be large); Tony can 
re-compile with _FFR_ZTAGS and then set "DiagnosticsDirectory" to be a 
location into which analysis is stored when verifications fail

If the canonicalizations are the same at both ends, there's a verification 
problem possibly involving key retrieval; if they're not the same, there's 
a problem with canonicalization and/or data modification in transit.

The fact that only one recipient is having that problem makes me think 
it's localized and thus more likely a transit problem of some kind.

One other possibility I can think of: Could there be an older public key 
in Tony's cache?

-MSK

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
dkim-milter-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dkim-milter-discuss

Reply via email to