Hi Fred,

Thank you for your information. Hope to see the implementation related paper
soon.

BTW, I like the idea of using GSE to realize NAT66 :)

Xiaohu 

> -----邮件原件-----
> 发件人: Fred Baker [mailto:[email protected]] 
> 发送时间: 2009年8月3日 18:58
> 收件人: Xu Xiaohu
> 抄送: [email protected]; 'IRTF RRG'
> 主题: Re: [rrg] Some concerns about ILNP
> 
> FYI - I had some students at Beijing University of Post and Telegraph
> (BUPT) do an implementation of GSE using several approaches 
> to this problem, one of which was ILNP. The key thing to the 
> study was an agreed-to-be-weird routing scenario that forced 
> a session to oscillate among the available address pairs on 
> every exchange; see section 1.4.1 of the attached for the 
> scenario. They tried kind of "all of the above" in terms of 
> endpoint identifier approaches, and they all failed with two 
> exceptions, GSE as described by Mike and ILNP. HIP couldn't 
> set up the association as it depends on the first few 
> messages getting across without the address changing until 
> the HIT is established, and this changed them there.
> 
> I would support RJA's contentions here; ILNP gets the job 
> done, even in the presence of EID duplication in the network.
> 
> The BUPT kids will be publishing a paper out of this, 
> assuming they can get interest from an appropriate journal.
> 
> 

_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to