Re: [Bier] Direction: draft-xie-bier-ipv6-encapsulation

[Michael McBride <[e-mail]>]

Hi Greg,

>To date, there has been insufficient support on the list to adopt the draft.

We won’t know unless we ask the WG.

>It is the BIER WG position at this time that a review by the 6man WG should take place before any further discussion in the BIER WG.

You mean the bier chairs position. Either way a punt to 6man for bierv6 work is fair. We will return and report.

>6MAN WG review aside, the draft still raises significant technical concerns within the BIER WG. Adding replication information into a programmatic header does NOT add replication to programmatic forwarding path, nor programmatics to your replication forward path. It is at this point where replication begins that all prograrmmatics cease due to the multi-point nature of replication.  We have a nearly three decades of experience trying to use unicast semantics to describe multi-point forwarding, and sometimes learning from the past requires leaving it.

Perhaps others can help me understand what this means. Regardless, what we will do for the time being, in bier wg, is continue working on gaining bier wg consensus of the requirements, in our draft-ietf-bier-ipv6-requirements, to help steer the wg toward agreeable solutions.

mike


Thanks you,
Chairs