ietf-dkim
[Top] [All Lists]

Re: [ietf-dkim] LWSP in base64-encoded public key TXT RR

2007-03-09 08:12:18
Mark Martinec wrote:
What would actually be better here, IMO, is to find out _how_ this
happened. That is, is this a pathological case for a DNS server we
ought to care about.

A bit of experimenting with BIND and reading the zone file syntax
in rfc1035 can explain what probably happened. All entries below are
syntactically correct as far as the zone syntax and DNS is concerned.

right._domainkey  TXT  ("k=rsa; p=MIG"
                        "MAG0udE82Yzb7"
                        "xLv4QnQIDAQAB")

wrong._domainkey  TXT   "k=rsa; p=MIG\
MAG0udE82Yzb7\
xLv4QnQIDAQAB"

wrong2._domainkey TXT  "k=rsa; p=MIG\013\010MAG0udE82Yzb7xLv4QnQIDAQAB"


This is actually why I thought this might be useful: if we're going to
give any informative advice about this at all, it would be useful to
couch it in terms of potential gotchas since the key sizes are long.
Just warning that the ABNF is a holdover from the signature header
doesn't give any motivation of why somebody would care about this.

                Mike
_______________________________________________
NOTE WELL: This list operates according to http://mipassoc.org/dkim/ietf-list-rules.html

<Prev in Thread] Current Thread [Next in Thread>