Re: [netmod] WG Last Call: draft-ietf-netmod-rfc7277bis-00

[Martin Bjorklund <[e-mail]>]

Hi,

Vladimir Vassilev <[e-mail]> wrote:
> On 12/16/2017 10:46 AM, Martin Bjorklund wrote:
> 
> > Hi,
> >
> > Vladimir Vassilev <[e-mail]> wrote:
> >> On 12/13/2017 04:26 PM, Vladimir Vassilev wrote:
> >>> Hi,
> >>>
> >>> On 12/13/2017 03:47 PM, Martin Bjorklund wrote:
> >>>
> >>>> Hi,
> >>>>
> >>>> Thanks for reporting this.  I'll add the missing origin.  But why did
> >>>> you think forwarding and mtu should be removed?
> >>> 1. IMO since <mtu> is not present in the <ipv4> container in the
> >>> Appendix A (<get-config>) example and does not have default value in
> >>> the model I still think it should be removed.
> >> Alternatively the ipv4/mtu node can be a good example of a
> >> origin="or:system" configuration.
> > Yes.
> >
> >>>>     In fact, I think I
> >>>> missed <enabled>,
> >>> 2. IMO both fixes adding <enabled> or removing <forwarding> should be
> >>> OK depending on the RFC6243 defined with-defaults capability
> >>> 'basic-mode' parameter advertised by the server. I was running the
> >>> example with basic-mode=explicit
> > Right.  I now have this:
> >
> >        <!-- other parameters from ietf-interfaces omitted -->
> >
> >        <ipv4 xmlns="urn:ietf:params:xml:ns:yang:ietf-ip">
> >          <enabled or:origin="or:default">true</enabled>
> >          <forwarding or:origin="or:default">false</forwarding>
> >          <mtu or:origin="or:system">1500</mtu>
> >          <address>
> >            <ip>192.0.2.1</ip>
> >            <prefix-length>24</prefix-length>
> >            <origin>static</origin>
> >          </address>
> >          <neighbor or:origin="or:learned">
> >            <ip>192.0.2.2</ip>
> >            <link-layer-address>00:01:02:03:04:05</link-layer-address>
> >          </neighbor>
> >        </ipv4>
> >        <ipv6 xmlns="urn:ietf:params:xml:ns:yang:ietf-ip">
> >          <enabled or:origin="or:default">true</enabled>
> >          <forwarding or:origin="or:default">false</forwarding>
> >          <mtu>1280</mtu>
> >          ...
> >
> > Do you think this is ok?
> Yes. The or:default data makes the example even better.
> 
> 1. However there is one more default value missing
> (/interfaces/interface[name='eth0']/enabled) for the example to be
> consistent

Note that the example has:

   <!-- other parameters from ietf-interfaces omitted -->

This covers the 'enabled' leaf (and more).  The idea is to let this
document focus on the ip parameters.

> 2. ... and in the last diff I unintentionally omitted the get-data
> output node <data> required namespace addition (this is also
> applicable to draft-ietf-netmod-7223bis-01):

Good catch!  I have applied this in both documents.

I also realized that the drafts don't use text from the latest
Security Considerations template; I have fixed that as well.


/martin




> 
> diff -u before2.xml after2.xml
> --- before2.xml    2017-12-18 11:41:54.029279321 +0100
> +++ after2.xml    2017-12-18 11:36:25.973850340 +0100
> @@ -1,4 +1,4 @@
> -     <data>
> +     <data
> xmlns="urn:ietf:params:xml:ns:yang:ietf-netconf-datastores">
>         <interfaces
>             xmlns="urn:ietf:params:xml:ns:yang:ietf-interfaces"
>             xmlns:ianaift="urn:ietf:params:xml:ns:yang:iana-if-type"
> @@ -7,6 +7,7 @@
>           <interface or:origin="or:intended">
>             <name>eth0</name>
>             <type>ianaift:ethernetCsmacd</type>
> +           <enabled or:origin="or:default">true</enabled>
>             <!-- other parameters from ietf-interfaces omitted -->
>             <ipv4 xmlns="urn:ietf:params:xml:ns:yang:ietf-ip">
>               <enabled or:origin="or:default">true</enabled>
> 
> 
> Vladimir
> >
> >
> > /martin
>