Re: [Bier] [babel] About bierin6 and signalling of encapsulation

Hi Juliusz,

1. statically configured (e.g. a given BIER domain is statically
     configured to use a given encapsulation);
Sandy: Yes, static configuration provides a simple way to implement bier forwarding.

  2. carried by a specific signalling protocol (call it the BIER Hello
     Protocol for now);
Sandy: Bier node information is advertised by IGP/ BGP protocol (as you mentioned "Bier hello protocol"), Bift-id is the key to find the corresponding forwarding item. But there's no restriction for the out-layer encapsulation. In fact, the Bier packets can be executed correctly whatever the out-layer is, such as Mpls, Ethernet or IPv6, and so on. The out-layer encapsulation is only used to make sure the Bier packet can be received correctly by the next hop router.
To your example, A sends the bier packet with whatever encapsulation, indeed the packet will be received and executed correctly by B. A should choose a reasonable encapsulation method though it can choose the method optionally. In fact there's consensus in one bier domain.

  3. carried by the routing protocol (e.g. carried by a sub-TLV of the IHU
     TLV in Babel, or using RFC 5613 signalling in OSPF).
Sandy: You needn't do that. But you can try if you insist on it.
________________________________
发件人: babel <babel-bounces@ietf.org> 代表 Juliusz Chroboczek <jch@irif.fr>
发送时间: 2017年11月17日 1:16:49
收件人: Tony Przygienda
抄送: bier@ietf.org; Babel at IETF
主题: Re: [babel] [Bier] About bierin6 and signalling of encapsulation

> you simply pick either one, it's a hop-by-hop decision.

What if A implements both encapsulations, and B only implements bierin6?
If A picks the native encapsulation, you've created a blackhole, no?

-- Juliusz

_______________________________________________
babel mailing list
babel@ietf.org
https://www.ietf.org/mailman/listinfo/babel