--On Wednesday, 11 April, 2007 12:07 +0100 Tony Finch
<dot(_at_)dotat(_dot_)at> wrote:
On Tue, 10 Apr 2007, william(at)elan.net wrote:
How would you see it working with extension ID?
Client sends KEEPALIVE <interval> at the start of the session
(or again later to change the interval). Server sends 1yz
reply after interval if normal reply is taking longer. Client
responds to 1yz with CONTINUE or RSET. This screws up
pipelining so the server probably should not offer pipelining
and keepalive, or perhaps the keepalive command turns off
pipelining.
Yep. I think it is also possible to model this as slightly
server-controlled with client confirmation, as outlined in one
of my earlier notes. But all of them have, I think, the
properties that:
(1) There is some specific negotiation/ agreement that
overrides the timeouts that would be exhibited by a 2821
implementation with no extensions advertised or
requested.
(2) The model of "send a command, get a unitary and
homogeneous (even if multi-line) reply" is not altered
through some back door that some 2821-conforming
implementations may not observe (or even notice).
john