[opendmarc-dev] Bug #8: Forgiving errant "\" characters

Benny Pedersen me at junc.org
Fri Aug 3 08:01:05 PDT 2012


Den 2012-08-03 16:15, Murray S. Kucherawy skrev:

> This might be an option somewhere down the line, but for now keeping
> them as separate modules while the protocol is in some flux is
> probably the better option.

yes, but since nearly all milters i know have lib in mind, there should 
be no problem use these libs as plugin in another milter, this keeps it 
simple from mta side

> Also, opendkim is a very large beast in
> terms of its internal complexity, and I'd hate to have to duplicate
> all of that functionality.

understandeble, i just think outside this scope for what the lib api 
can do later, before its to late to reduce duplicates in lib api

>> as it is now we end using 100 milters that is not working well 
>> together :(
>
> I'm not having any problems running sid-milter, opendkim, and
> opendmarc in a chain.  Is anyone else having troubles that haven't
> been reported?

sid-milter reject domains without spf records

it seems not tested with non default startup options :(

dig lists.isc.org spf
dig lists.isc.org txt

where is valid spf header ?

>> and for dmarc it olso stops forged headers if opendmarc just test 
>> headers on recieve ?
>
> Sorry, I don't understand what you're suggesting here.  Can you give
> an example?

i readed you saying that opendmarc just checking headers not try to 
validate the headers opendkim/sid-milter  adds this could open a 
potiential problem in forged headers, render opendmarc to not be stable



More information about the opendmarc-dev mailing list