[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Thoughts on NetConf Requirements



>>>>> David T Perkins writes:

David> Given this, you could send the XML as text, encode it in
David> binary, put it directly on top of TCP, or on SSH, or on
David> TLS. You could use one transport connection per channel, or use
David> BEEP to provide support of multiple channels over one transport
David> connection.  You might be able to get by with a single channel
David> (if everything works OK). But you might need a channel to abort
David> a request or its response. 

I have been told by CMIP implementors that in old CMIP land, the
cancel-get primitive was a real pain to implement right and in
practice hardly needed since in those cases where you really had to
cancel a request, simply closing the underlying transport connection
was much cheaper to get done right (and an implementations has to
handle this case anyway).

David> And another channel to receive the events and/or tracing that
David> starts to happen because of the command that you sent on the
David> operations channel.

The EPP folks seem to have solved this by having kind of a message
queue on the server which you simply poll to retrieve pending
messages. (Sure, this raises issues about how to deal with message
buffers if the client does not poll frequently enough or the latency
issue if you only poll every say 5 minutes. Anyway, I do not really
propose to use EPP as a template here nor do I want to go into a
discussion of push vs. pull of asynchronous messages here.

But I have questioned the need for BEEP as the required transport in
the past because I am not totally convinced that having multiple
channels over a single transport connection is indeed a requirement
for netconf. And operators told us several times that they want to be
able to type (or cut n paste) messages into a (secure) transport
connection to handle some special situations.

BTW, also from this perspective, I find the EEP approach of using XML
(every protocol message is a separate XML document) easier and more
attractive.

/js

-- 
Juergen Schoenwaelder		International University Bremen
Phone: +49 421 200 3587		P.O. Box 750 561, 28725 Bremen, Germany
Fax:   +49 421 200 3103		<http://www.iu-bremen.de/>

--
to unsubscribe send a message to netconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>