Re: [Anima] 答复: Request for presentation

[Michael Richardson <[e-mail]>]

I've only very briefly scanned
   https://tools.ietf.org/html/draft-li-rtgwg-protocol-assisted-protocol-03

Brian E Carpenter <[e-mail]> wrote:
    > I don't really understand why you might need a new protocol for
    > this. It sounds like a very good application for the ANIMA model, and
    > everything you describe could easily be expressed in GRASP objectives
    > (synchronization or negotiation are both covered). You discuss this in
    > Section 7.

I'd split this up in three points:

1) It seems like a very good use for the ACP.
   If you are debugging a protocol problem, then you likely have some part of
   the network that has reachability issues.

2) I think that the ACP eliminates the need for PAP to be peer to peer, or
   for it to be UDP.
   It should be between the NOC and the router, not between routers.

3) GRASP CLEARLY has a place in discovering the routers which speak this
   protocol, and/or in having the routers discover the PAP collector in the NOC.


Whether or not GRASP is the right way to transfer the data itself, I'm not
sure.  I'm unclear if access to the raw routing protocol frames (OSPF, BGP,
etc.) is desired, or if some higher-level debug is envisioned.

If raw protocol, then it seems like a span/mirror/ functionality using IPFIX
or PCAPNG [over ACP] would be appropriate.

If you want higher-level diagnostics, then I would use RESTCONF and YANG.

--
Michael Richardson <[e-mail]>   . o O ( IPv6 IøT consulting )
           Sandelman Software Works Inc, Ottawa and Worldwide