Re: event classes (sec. 3.5)

[Andy Bierman <ietf@andybierman.com>]

Phil Shafer wrote:
> Balazs Lengyel writes:
> > Does this mean that for each operational command you have a separate RPC/operation ?
>
> Yup, under junoscript and netconf, each junos command is available
> as a distinct RPC.
>
> > Is the below example the complete XML doc or is there 
> > still some RPC wrapper around it ?
>
> What I sent was the complete RPC, as it would appear inside the
> netconf <rpc> element.  Here's another example:
>
>         <get-interface-information>
>           <terse/>
>           <interface-name>ge-0/1/2</interface-name>
>         </get-interface-information>


I like special RPCs for this kind of thing.
IMO this is a different sort of RPC than the type in the
Notifications draft.   My biggest concern is not a 1000
little APIs to deal with, but rather multiple APIs that
deal with the "netconf write model" or data models
in inconsistent ways.

We have a "write model" based on 3 standard conceptual data-stores
(i.e., <candidate>, <running>, <startup>) and a set of operations
to deal with various complexities with the native devices.
IMO, defining new RPCs which modify NV-saved config data is wrong,
if they ignore or contradict this model.



> All our RPCs are documented under www.juniper.net/support/junoscript/.
> All these RPCs are available under netconf as well, since we drive
> the command, file, patch, xml, junoscript, and netconf parsing off
> the same meta-data.
>
> Thanks,
>  Phil

Andy

--
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/>