Re: [OSPF] Fwd: I-D Action:draft-venkata-ospf-dynamic-hostname-02.txt

[Danny McPherson <danny at tcb.net>]

On Jan 29, 2008, at 10:22 PM, Nitin Kakkar wrote:

> Danny,
>   I have couple of comments
>   1) IMHO we need to explicitly define flooding scope for RI
> information,

What explicitly beyond the current text were you hoping to see?

   The flooding scope is controlled by
   the Opaque LSA type in OSPFv2 and by the S1 and S2 bits in OSPFv3.
   For area scope, the Dynamic Hostname TLV MUST be carried within an
   OSPFv2 Type 10 RI LSA or an OSPFv3 RI LSA with the S1 bit set and S2
   bit clear.  If the flooding scope is the entire routing domain (AS
   scope), the Dynamic Hostname TLV MUST be carried within an OSPFv2
   Type 11 RI LSA or OSPFv3 RI LSA with the S1 bit clear and the S2 bit
   set.  An AS boundary router (ASBR) may choose to send an AS scope
   Dynamic Hostname TLV, whereas area boundary router (ABRs) and
   internal routers may choose to send an area scope Dynamic Hostname
   TLV.  If a router originates Dynamic Hostname TLVs with an IGP  
domain
   (AS) flooding scope, it MAY omit sending area-scoped Dynamic  
Hostname
   TLVs.

>   2) We need to define selection/priority rules when multiple  
> hostnames
> map to a single router-id (again tagged to flooding scope). IMHO it is
> possible that some routers have same router id within AS/Area/Link.

Really? RFC 2328:

        Router ID
            A 32-bit number assigned to each router running the OSPF
            protocol.  This number uniquely identifies the router  
within
            an Autonomous System.

Am I misinterpreting your comment here?


> On the side note, During Chicago IETF there were some
> discussions/concerns about impact of non routing information on ospf
> convergence. And one of the proposals was separation of routing & non
> routing information.
> I think Now is the time to put a process in place, IMHO In  
> documents we
> can explicitly specify TLV's as routing or non routing, Cases like RI
> are fairly obvious, but some where down the road we might have
> extension's that may not be so obvious, so its better to mark tlv as
> routing/non routing.

While I agree, I'm not sure there's anything we explicitly need to
do here, in particular given that we're employing an RI Opaque
LSA for this purpose.  Do you have something specific in mind?

> Just my 2cents, I am sure Acee & other can chip in with their $$

Thanks for the feedback!

-danny

_______________________________________________
OSPF mailing list
OSPF at ietf.org
https://www1.ietf.org/mailman/listinfo/ospf