RE: Where do we go from here?

["Ken Crozier" <kcrozier@cisco.com>]

The real meat of this problem is the transport, the messaging, the security
(which nobody has mentioned up to now) and application behavior. XML is just
how we define some variables and should be regarded as such.

So maybe if we start separate discussions on the following, then maybe we
can get somewhere..

Transport....
What are the base requirements for the transport ?
Do we need multiple channels to each device ? Say a data channel and a
command channel
Do we need sequence numbers ?
Do we need retires ?
Does it have to be persistent i.e. connected all the time ?

Messaging protocol
What are the base requirements ?
What verbs do we need ? (reload, new config, parse & execute, parse & store,
loose your brains for 5 minutes and just come back, lock the config, new
software image, erase flash)
What return codes do we need ?

Security
All this needs to be done securely
Is it good enough to have trusted domains between the server & the client,
or must each and every command be authorized by the end user.
Do we need to authenticate each and every packet
How do we authenticate the server and client
Do we care about certificates

Application behavior
Transaction oriented
Role back commands.
Artificial intelligence in the network. i.e. the command that was just sent
down to one of the core routers with 'no ip routing' is stupid and i will
not execute it

There is work that needs to do be done in this area. Either the IETF does it
or other groups will.
ken crozier

-----Original Message-----
From: owner-xmlconf@ops.ietf.org [mailto:owner-xmlconf@ops.ietf.org]On
Behalf Of RJ Atkinson
Sent: Friday, August 09, 2002 9:39 AM
To: brunner@ccrle.nec.de
Cc: xml list
Subject: Re: Where do we go from here?



On Friday, August 9, 2002, at 05:02 , Marcus Brunner wrote:
>> I am quite confident that there is no single magic set of operator
>> requirements that applies to all operators.  I speculate that there
>> is a lot of commonality in operator requirements.  Finding and focusing
>> on that commonality seems useful, if a bit of a challenge.  My
>> experience
>> so far is that enterprise operators often have a very different view
>> of the world than (for example) large core backbone operators.
>
> I agree that there is not a single set of operator requirements.
> However,
> it seams that the IETF did not do a good job in at least partly fulfil
> some requirements of some operators (for config mgt).

As near as I can tell, the IETF hasn't produced widely deployable
configuration management technology or standards at all so far.

Ran


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


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