On May 1, 2014:3:08 PM, at 3:08 PM, Andy Bierman <[email protected]> wrote:

> 
> 
> 
> On Thu, May 1, 2014 at 11:28 AM, Joel M. Halpern <[email protected]> wrote:
> Jeff, I think the problem is subtler.
> Even if I2RS wants to manipulate data represented in the config-true (CLI 
> equivalent) tree, the I2RS data is still actually separate.  If you ask what 
> the CLI has set, you get one answer.  If you ask what I2rS has set you get a 
> different answer.  And if you ask what is actually operating you get the 
> result of the arbitration.
> At least that is what the architecture describes.
> 
> This clearly can be modeled in YANG.  It may take a little more effort.
> 
> 
> I didn't think the I2RS charter covered writing to the local configuration,
> that is maintained by NETCONF.  I hope not.
> 
> But if there is some intent to support a "copy-to-local-config" option
> in the I2RS protocol, then a mapping from the I2RS data to the
> local config data would be needed.

        Right. The thinking was to include some option in the protocol 
extensions to trigger a copy-to-local-config "later" either of the entire 
running config, or parts of it. But the normal operation was only (and as 
quickly as possible) to the running config/rib/system.

        --Tom


> 
> 
> Yours,
> Joel
> 
> Andy
>  
> 
> On 5/1/14, 2:17 PM, Jeffrey Haas wrote:
> Andy,
> 
> On Thu, May 01, 2014 at 11:10:11AM -0700, Andy Bierman wrote:
> that is not what the "edit-data" extension does.
> It is a sub-statement that identifies I2RS editable data no matter what
> YANG module is used to define the data.  The YANG config=true statement
> identifies a data node as part of the configuration datastore. The edit-data
> extension cannot be used on a config=true node, only non-configuration.
> 
> module foo {
>     ...
>     import i2rs { prefix i2rs; }
> 
>     container config-data {
>         // NETCONF definitions here
>      }
> 
>      container i2rs-data {
>         config false;
>         i2rs:edit-data;
>      }
> 
> You're crystal clear to me now. Thank you.  This was not permitting I2RS to
> be able to interact with/overlay part of config-data (per your example) as I
> had hoped.
> 
> Nesting is also possible -- up to the data modeler to decide what to use:
> 
>     1) i2rs-data inside config-data
>     2) i2rs-data inside oper-data
>     3) oper-data inside i2rs-data
> 
> What I had been hoping was for config-data to permit elements that
> were effectively config true,i2rs.
> 
> (Or s/i2rs/ephemeral/ as per previous messagees.)
> 
> IMO, I think we want multiple data stores for configuration for adding
> routes to the rib.
> 
> How many does I2RS need for itself?
> 
> To clarify, a config-data data store, an i2rs-data datastore where
> the i2rs portion may overlap.
> 
> Since I'm apparently not making my point well, I'll see if I can mockup my
> understanding in a netconf like transaction.  (For Dean, I haven't caught up
> in my reading to restconf, else I'd do that.)
> 
> -- Jeff
> 
> _______________________________________________
> i2rs mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/i2rs
> 
> 
> _______________________________________________
> i2rs mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/i2rs

Attachment: signature.asc
Description: Message signed with OpenPGP using GPGMail

_______________________________________________
i2rs mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/i2rs

Reply via email to