It appears that Richard Clayton <[email protected]> said: >-----BEGIN PGP SIGNED MESSAGE----- >Hash: SHA1 > > >As discussed "in Vienna" (see the minutes) > >At present there is no limit in the spec on the number of Message- >Instance or DKIM2-Signature header fields that can be present in a >message .... > >The only reason that there would be a lot of them is either because >someone is trying to waste your time or because a message is going round >in a loop. > >Detecting loops by counting Received: header fields is standard >practice... and there is no need to have more Message-Instance or >DKIM2-Signature fields than Received: header fields in your message
Urrgh. I hope you mean the limits should be similar. I can think of lots of reasons I'd have more Message-Instance than Received in a single message. Say if a message goes through a mailing list and the list software and the MDA mutate it twice on the same machine. >Anyway -- I think we should limit the number of M-I and D-S header >fields; and I would suggest 31 as a suitable limit (using a power of two >minus one to allow for fancy coding tricks if you like that sort of >thing). Seems reasonable. R's, John _______________________________________________ Ietf-dkim mailing list -- [email protected] To unsubscribe send an email to [email protected]
