ietf-nntp RFC977bis w.r.t. authentication

Stan Barber sob at academ.com
Tue May 5 16:39:56 PDT 1998


> As I posted earlier, I doubt it's 8bit enough.  Can it handle arbitrary CRLF
> sequences in protocol elements?  Can it handle NULs in protocol elements?

Good point. Propose some text to address the issue and then the rest of us
can review it. If you want to see this fixed, you have to contribute to the
process of fixing it.

> The point is that authentication protocols have a nasty tendency of
> generating a LOT of variible length binary data.  So you HAVE to define how
> the SASL data is wrapped.  I know of 2 ways of doing this used by other
> protocols.  LDAP BER encodes the binary data and transfers it that way
> (which is reasonable since it's an ASN.1 protocol).  The other way is to
> base64 the data and send the binary blobs as CRLF delimited strings (POP,
> IMAP, and SMTP-AUTH) do this.
> 
> Since NNTP is a text protocol I'd STRONGLY suggest the latter.

See above. Send proposed text for the draft.

-- 
Stan   | Academ Consulting Services        |internet: sob at academ.com
Olan   | For more info on academ, see this |uucp: mcsun!academ!sob
Barber | URL- http://www.academ.com/academ |Opinions expressed are only mine.



More information about the ietf-nntp mailing list