Hi,

Thanks for your comments. Please see inline below.

Cheers,
Ian

> On 8. Nov 2017, at 10:47, Andy Wingo <[email protected]> wrote:
> 
> Hi,
> 
> On Mon 30 Oct 2017 16:12, [email protected] writes:
> 
>>        Title           : A YANG Data Model for IPv4-in-IPv6 Address plus 
>> Port Softwires
>>        Authors         : Qi Sun
>>                          Hao Wang
>>                          Yong Cui
>>                          Ian Farrer
>>                          Sladjana Zechlin
>>                          Mohamed Boucadair
>>                          Rajiv Asati
>>      Filename        : draft-ietf-softwire-yang-02.txt
>>      Pages           : 41
>>      Date            : 2017-10-30
> 
> I had a look at this for the Snabb lwAFTR and found a few issues.
> 
> 1. "hairpin-ipv4-packets" and "hairpin-ipv4-bytes" need to be state
>    data, not config data.  Consider moving them inside the
>    "traffic-stat" container.

[if - This was an oversight when we moved from ‘config false’ at the leaf level 
to the container level. The following change fixes it:

container traffic-stat {
              config false;
              description
                “traffic-stat”;
              uses softwire-common:traffic-stat;
              leaf hairpin-ipv4-bytes {
                type yang:zero-based-counter64;
                description “IPv4 packets locally routed between two CEs
                  (hairpinned).“;
              }

             leaf hairpin-ipv4-packets {
                type yang:zero-based-counter64;
                description “IPv4 bytes locally routed between two CEs
                  (hairpinned).“;
              }

             leaf active-softwire-num {
                type uint32;
                description
                  “The number of currently active softwires on the
                  lw4over6 (binding) instance.“;
              }
            }

I’ve also noticed that config false hasn’t been set for the algorithm traffic 
stats. I’ll fix that as well.

> 
> 2. "dropped-ipv4-bytes", described as counting bytes from dropped IPv4
>    fragments, should probably be named "dropped-ipv4-fragment-bytes" or
>    something.  Note there is already "dropped-ipv4-byte" (sic) for
>    dropped IPv4 bytes as a whole.

[if - will fix]

> 
> 3. "dropped-ipv4-bytes" (from (2)), "dropped-ipv4-fragments",
>    "hairpin-ipv4-packets" (from (1)), and "hairpin-ipv4-bytes" (from
>    (1)) are plural whereas the rest of the counters are singular.  One
>    or the other should probably be chosen.

[if - Will fix]

> 
> 4. The type of "icmpv6-errors-rate" is "uint16".  This is probably too
>    restrictive.

[if - I’ll up it to uint32]

> 
> Regards,
> 
> Andy
> 
> _______________________________________________
> Softwires mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/softwires

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

Reply via email to