Peering with other pTLA

[Bob Fink]

At 10:53 AM 3/1/99 +0800, Yann-Ju Chu wrote:
>	We have bocame a pTLA site in 6Bone and are trying to build peering with 
>other pTLAs. I have studied two drafts: "Multihomed routing domain issues 
>for IPv6 aggregatable scheme" and "6Bone routing practice", but still are 
>confused with something.
>	1. Should I normally announce only my own address prefix (3ffe:3600::/24) 
>to my peers?

In the spirit of the 6bone you should be willing to provide transit to all
other 6bone pTLAs (see section 4 of the 6Bone Routing Practice I-D below)
thus you can/should be willing to announce other pTLAs as well as your own.


>	2. If the answer is right => I should build full mesh connection with all 
>6Bone pTLAs, if I want a complete access to all 6Bone sites?

Given the above, you do not need to build a full mesh, which is why we
recommend 3 (again, see excerpt below).


Bob

===
draft-ietf-ngtrans-6bone-routing-01.txt 6Bone Routing Practice 1 June 1998

4 Routing policies

6Bone backbone sites maintain the mesh into the backbone and provide an as
reliable as possible service, granted the 6Bone is an experimentation tool.
To achieve their mission, 6Bone backbone sites MUST maintain peerings with
at least 3 (three) other back bone sites.

The peering agreements across the 6Bone are by nature non-commercial, and
therefore SHOULD allow transit traffic through.

Eventually, the Internet registries will assign other TLAs than the 6Bone
one (currently 3FFE::/16). The organizations bearing those TLAs will
establish a new IPv6 network, parallel to the 6Bone. The 6Bone MIGHT
interconnect with this new IPv6 Internet, but transit across the 6Bone
will not be guaranteed. It will be left to each 6Bone backbone site to
decide whether it will carry traffic to or from the IPv6 Internet.

-end