Re: [pim] draft-ietf-pim-yang-02 WGLC

[Stig Venaas <[e-mail]>]

Hi

I now took some time to review the draft more carefully. Here are
comments in addition to what I posted earlier for the WGLC.

First some potential issues with the models.

I see for hello-interval, jp-interval etc. the data type is
timer-value which can also be infinity and no-expiry. Isn't
this type to be used for something counting down? These intervals
would always be some number of seconds and won't change unless the
configuration changes.

Also I see the PIM-SM spec calls them periods, like hello-period
and jp-period. So maybe those names would be better?

I also see "expiration" is usually timer-value, but I see for
the multicast-route-attributes we have expiration that is uint32.
2.1

In the BSR model I see:

leaf hash-mask-length{
           type uint8 {
             range "0..32";
           }

The length can be larger for IPv6.

It might be useful to add some text explaining what the multipliers do.
I know some implementations use this concept, but it is not in any IETF
documents that I know. So good to explain it somewhere in this document.


The issues below are more minor.

   The model covers PIM Sparse Mode [RFC4601], including the Source-
   Specfic subset [RFC3569], Dense Mode [RFC3973], and Bi-directional
   PIM [RFC5015].

Typo "Specific"

   The PIM extensions represented in the model include BSR [RFC5059] and
   Anycast RP [RFC4610].

I think it should say "Anycast-RP".

The language in these paragraphs in 2.1 seems to say that this is still
work in progress, and should probably be changed a bit. The last
paragraph is mostly ok, but I think it should say "specified in separate
documents", rather than "future Internet Drafts".

2.4 starts with the words "The current draft". It should perhaps say
"This document".

What to do with section 3? You probably should change the language a
bit, and also not have a TODO list. You could perhaps make it talk about
the scope of the model, limitations, and what is left out?

In 4.1 you need to replace "our current direction" with some other
language. Maybe some of the text can be removed.

In 4.3 it says "ipv6-anycast-rip" in 2 places. Also 2 places it says
"range-poligy".

In 4.5 I see a couple of "offer-multipler".

For all these typos, please search and look for additional occurrences.
I see "multipler" several places in 5.1 too.

In 5.1 it says a couple of places"configuraiont". Also in description
of DR priority, perhaps capitalize DR.

Also I see a "list neighbor-ipv4", where it should be ipv6.

In the descriptions it says "address family" in some places, and "af"
other places. Maybe try to make it consistent. Consider "AF" instead of
"af"?

"The number of bsr messages" should have BSR capitalized.

Some places in 5.2 it says "Gouring".

I think it should say "IPv4" and "IPv6" in the descriptions, rather
than "ipv4 and ipv6". I see both used.

In 5.3, I think you should capitalize RPT, SPT and DR in the
descriptions. Also I see a description where it says "Spars Mode". I
see various ways of writing IPv4 and IPv6 in the descriptions. I think
they should be written like I just did. It says "Gouping" a few places.

In the description of passive it says:
            "Specifies that no PIM messages are sent out of the PIM
             interface, but the interface can be included in a multicast
             forwarding entry.";
Doesn't it also mean that no PIM messages are accepted?

In 5.5 there are two descriptions ending with "..". I also see some of
the same issues as mentioned above. Capitalization and typos like
"gouping" and "multipler".

Section 7
Please try to think if there could be any issues.

Section 8
I'm not sure if there are IANA considerations. If there are none, then
write that.

The 4601 reference should be updated to RFC 7761.

Stig