ietf-nntp AUTHINFO SASL protocol choices

[Russ Allbery]

Charles Lindsey <chl at clw.cs.man.ac.uk> writes:

> It would require great skill to write an implementation that would
> accept 1024 octets for the BLAH command, but would fail to accept 1024
> octets for all other commands.

Not really; it would just require a poor design.  :)  Think, for example,
of an old server onto which was tacked a SASL authentication handler that
was completely separte; when it gets the AUTHINFO command, it hands off
the connection to the SASL handler, which does its thing, sets some
variables, and then hands back to the rest of the code.

It's quite possible for some implementation strategies to result in that
situation, but those aren't implementation strategies that I'd recommend
to anyone.

-- 
Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>