[Lsr] 答复: Is it necessary to expand the IS-IS level to 8?

["Aijun Wang" <[e-mail]>]

Hi, Les:

 

We just want to distinguish the passive interfaces from other normal
interfaces within ISIS domain.  It seems that the ¡°Attribute Flags¡± that
described in https://tools.ietf.org/html/rfc7794#section-2.1 is the most
appropriate place to extend to carry such information.

If so, we can write one draft to define one more attribute flag to
accomplish this.

 

Or, is there any other way to fulfill this task? Originally, I want to reuse
the reserve bits before ¡°Circuit Type¡± field.  It seems not the right
direction. 

 

And on the other hand, occupying all the reserved bits for the unnecessary
expansion is also not the right direction? 

 

Best Regards.

 

Aijun Wang

China Telecom

 

·¢¼þÈË: [e-mail] [mailto:[e-mail]] ´ú±í Les Ginsberg
(ginsberg)
·¢ËÍʱ¼ä: 2020Äê1ÔÂ6ÈÕ 12:52
ÊÕ¼þÈË: Aijun Wang; [e-mail]
Ö÷Ìâ: Re: [Lsr] Is it necessary to expand the IS-IS level to 8?

 

Aijun ¨C

 

Regarding the number of levels, Tony responded to a similar comment several
months ago ¨C please see
https://mailarchive.ietf.org/arch/msg/lsr/dKJBcU59TNuY3rxgjd6iXIq6sYg

 

<snip>

> It seems not necessary to have 8 levels of hierarchies. 3 or at most 4
levels of hierarchies should be enough. IS-IS with 3 levels of hierarchies
may support a network with 1k*1k*1k nodes, which is about 10^9 = 1 billion
nodes. IS-IS with 4 levels of hierarchies may support a network with
1k*1k*1k*1k nodes, which is about 10^12 = 1 trillion nodes.

 

 

This is correct.  It¡¯s not absolutely necessary. However, as Robert
mentioned, it does give the network designer flexibility to create the
hierarchy that matches the needs of his network.  The cost of the additional
levels is very small, given that we¡¯re considering adding any levels at
all, so it seemed sensible to define all of the levels at once.

 

¡°From an architectural point of view, if a number isn¡¯t obviously too
large, then it¡¯s probably too small.¡±  ¡ª Ross Callon

 

<end snip>

 

Link type in OSPF is not at all the same thing as level in IS-IS ¨C so I do
not find that analogy appropriate:

 

>From https://tools.ietf.org/html/rfc2328 page 128

 

<snip>

Link type   Description       Link ID

                   __________________________________________________

                   1           Point-to-point    Neighbor Router ID

                               link

                   2           Link to transit   Interface address of

                               network           Designated Router

                   3           Link to stub      IP network number

                               network

                   4           Virtual link      Neighbor Router ID

 

 

                           Table 18: Link descriptions in the

                                      router-LSA.

<end snip>

 

   Les

 

 

From: Lsr <[e-mail]> On Behalf Of Aijun Wang
Sent: Sunday, January 05, 2020 8:03 PM
To: [e-mail]
Subject: [Lsr] Is it necessary to expand the IS-IS level to 8?

 

Hi, Tony, Les and Paul: 

 

As I read the draft
https://tools.ietf.org/html/draft-ietf-lsr-isis-extended-hierarchy-00, and
notice the proposal to expand the reserved bits in ¡°Circuit Type¡± to cover
the level 1-8 in ISIS domain.

Here I just want to know is it necessary to expand the IS-IS level to 8 and
occupy all the reserved bits in the PDU format?

 

As compared with the OSPF format, there is one field to describe the ¡°Link
Type¡± (5 bits, currently define only four types). We want to use the
reserved bits in current ¡°Circuit Type¡± of ISIS PDU format to fulfill the
similar tasks. 

 

Can this draft leave at least 2 reserved bits for this purpose? There are
many ways to tackle the scale of networks, who will design their network in
level 8 hierarchy? As I estimated, Level 4 may be the acceptable highest
network hierarchy.

 

 

Best Regards.

 

Aijun Wang

China Telecom