Re: IS-IS NSAP address

rthakker · ‎12-08-2005

Friends,

I am bit confused about IS-IS NSAP addressing and need some clarification.

* Addresses starting with value 49 (AFI = 49) are considered as private addresses; Addresses starting with AFI values 39 and 47 represent ISO Data Country Code and ISO International Code Designator, respectively.

* The last byte is the N-Selector (NSEL) and must be specified as a single byte length (preceded by a “.”). A NET definition must set the N-Selector to “00”.

This are some examples I have seen:

1. The NSAP 47.0001.aaaa.bbbb.cccc.00 consists of:

for IS-IS:

– Area = 47.0001

– System ID = aaaa.bbbb.cccc

– N-Selector = 00

for ISO-IGRP:

– Domain = 47

– Area = 0001

– System ID = aaaa.bbbb.cccc

– N-Selector = 00

2. The NSAP 39.0f01.0002.0000.0c00.1111.00 is regarded:

by IS-IS:

– Area = 39.0f01.0002

– System ID = 0000.0c00.1111

– N-Selector = 00

by ISO-IGRP:

– Domain = 39.0f01

– Area = 0002

– System ID = 0000.0c00.1111

– N-Selector = 00

I have seen some different address at client side and want to know if it’s valid or not and if yes then could you tell me if I am right or wrong?

89.0900.0211.2074.2220.2100

89.0900.1511.2074.2220.9600

89.0900.0111.2074.2222.1700

89.0900.0711.2074.2220.0100

n by IS-IS:

– Area = 89.0900.0211 (Can you use 89? or you must always use 49 ? if i will use 89 then any implications?

– System ID = 2074.2220.2100

– N-Selector = ?? (Missing don’t know why? Is it ok if you miss NSEL value or if it’s not specified then it’s considered by default to 00 ??? )

Plz help me understanding this

Thanks,

Rits

Georg Pauwen · ‎12-08-2005

Hello Rits,

did the NET addresses mentioned by you above actually work ? The n-selector of the NET must always be 0, and must be one octet in length. The entire NET address must also start with an octet. So anything with an n-selector other than 00 should not work. As with regard to the area id: it doesn´t matter what it is, as long as it starts with an octet...

Regards,

GP

rthakker · ‎12-09-2005

I was also surprised but as I mentioned this works now I don’t know how that’s what I want to know if it’s valid address or not ??

Georg Pauwen · ‎12-09-2005

Hello Rits,

it is not a valid address, but if it works, that of course makes my answer useless.

I would stick with the guidelines and not use anything else than what defines a valid address: valid addresses have the n-selector set to 0 and use one octet, and they also have to start with an octet. The length of the NET address is between 160 bits (20 bytes) and 64 bits (8 bytes).

Regards,

GP

rthakker · ‎12-09-2005

I have just seen very good example but i still don't understand why Nselect is not used in my case..

Example Cisco config of System ID

Interface Loopback 0

IP address 192.168.1.24

Router isis

Net 49.0001.1921.6800.1024.00

In Example, the loopback address is transformed as follows:

Step 1. Each octet in the dotted-decimal notation of the loopback IP address that is not three digits

is prefixed with zeros, padding it to three digits, as follows:

192.168.1.24 ---> 192.168.001.024

Step 2. After Step 1, you have 12 digits, which you can easily rearrange into three groups of 4

digits, as follows:

192.168.001.024 ---> 1921.6800.1024

Step 3. 1921.6800.1024 can then be used as the unique SysID of the router. The area prefix and

NSEL suffix are added to obtain the complete NSAP address, as shown in Example 4-6.

Example Obtaining the Complete NSAP Address

1921.6800.1024 ---> 49.0001.1921.6800.1024.00

------- -------------- --

^ ^ ^

| | |

AreaID SysID NSEL

This method of defining the NSAP or NET is frequently used by Internet service providers that run the ISIS

protocol as IGP. Even though this might not be the only intuitive way, the method offers operational

convenience by associating multiple applications to a single loopback address on the router. A single

loopback address can be used as the BGP router ID, the basis for the IS-IS SysID, the MPLS/TE router ID,

and for network management applications. OSPF also uses a loopback address on a router, when available,

as the router ID; in typical situations, however, OSPF and IS-IS are not used together as IGPs in the same network, even though that is possible for merging two originally separate networks or during a routing protocol migration from one to the other. Because the Area ID and the NSEL are the same for all nodes in the same area, it's the unique SysID that provides uniqueness to the NET of a router. An IS-IS router must have at least one NSAP configured, and the SysID must be unique for the area in which it belongs.

So my qiestion is still same is it ok to miss NSEL value or not ??

Any comments on this ??

Rits

raju_raghavan · ‎12-09-2005

Ritz,

The last octect is allways the system id irresective of the " . " delimiter.

So for 89.0900.0211.2074.2220.2100

SEl = 00

Sys ID = 1120.7422.2021

Area Address = 89.0900.02

You can see this from " sh clns protocols " command...

Hope this helps....Thanks

rthakker · ‎12-09-2005

This is cool.. Many Thanks Raju :-)

#sh clns protocol

IS-IS Router: 8900

System Id: 1120.7422.2021.00 IS-Type: level-2

Manual area address(es):

89.0900.02

Routing for area address(es):

89.0900.02

No interfaces in domain/area.

Redistribute:

static (on by default)

Distance for L2 CLNS routes: 110

RRR level: none

Generate narrow metrics: none

Accept narrow metrics: none

Generate wide metrics: level-1-2

Accept wide metrics: level-1-2

Rits