Hi Sarah (OSPF List copied),

See inline.

On Jan 27, 2013, at 2:34 PM, sarah boufelja wrote:


Dear Sir, I am a engineering student at Telecom Bretagne, a french graduate 
school and a research center in communications and computer science. I am 
studying in my project routing protocols and I found an Internet-draft that has 
been written by you and Mr. Jari Arkko about OSPFv3 auto-configuration. I am 
interested in this subject and I have some questions concerning the draft. I 
would be really thankful if you enlightened me:

1- My first question concerns the part 5.4 in the draft (Change to received 
self-originated LSA processing): it's written that the router, when it detects 
that some instances of its self-originated LSA are purged or reflooded, it's an 
indication of a possible router-ID duplication and it must delay 
self-originated LSA processing for LSA that have recently been purged or 
reflooded for ten seconds and an exponential back-off of 1 to 8 seconds for the 
processing delay. I don't really figure out this specification, what about the 
AC -LSA, isn't sufficient to detect and resolve any duplication ID issue?

Possibly the AC-LSA isn't the first LSA to be received by the router joining 
the OSPFv3 routing domain with the same OSPFv3 Router-ID as another router in 
the same domain. Hence, this is to mitigate any flooding wars until the OSPFv3 
routers realize that a duplicate router-id condition exists.


2- Concerning the IPv6 global addresses, the specifications introduced the 
AC-LSA that can be used in the future to carry some additional information 
about global IPv6 prefixes, can we envisage for instance to use this LSA to do 
a stateless autoconfiguration by giving the router a prefix to construct  a 
global unicast address without any human intervention?

There is a draft proposing just this in the HOMENET WG.

https://datatracker.ietf.org/doc/draft-arkko-homenet-prefix-assignment/


Thank you very much in advance for your help.

Thanks,
Acee




Cordially.

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

Reply via email to