Hi Jeffrey,

On Feb 15, 2012, at 10:23 AM, Jeffrey (Zhaohui) Zhang wrote:

> Acee,
> 
> Thanks for your comments. Please see below.
> 
>> -----Original Message-----
>> From: Acee Lindem [mailto:[email protected]] 
>> Sent: Friday, February 03, 2012 12:45 PM
>> To: [email protected] List
>> Subject: [OSPF] Comments on OSPF Hybrid Broadcast and P2MP 
>> Interface Type - draft-ietf-ospf-hybrid-bcast-and-p2mp-00.txt
>> 
>> I've reviewed the subject document and have a couple comments. 
>> 
>>    This document introduces some new dependencies since 
>> routers connected to the hybrid network are using the DRs 
>> Router-LSA to originate their own Router-LSA. Hence, if some 
>> goes wrong with the distributed DR election algorithm, 
>> connectivity will fail differently than before. Is anyone 
>> concerned about this? If so, we could add a requirement that 
>> the non-DR router and the DR agree on who is DR prior to the 
>> non-DR router using the DR-Router's Router-LSA to originate its own. 
> 
> Makes sense. How about the following?
> 
>   o  If a router is not the DR and has a full adjacency to the DR,
> -->   and both the DR and this router agree on the DR role, it
>      MUST add a Type 1 link corresponding to each neighbor that is in
>      state 2-Way or higher and to which the DR's router LSA includes
>      a link.

That's fine. Hope you guys agree with the additional verification. 

> 
>> 
>> 
>>    The document needs a "Management Considerations" section 
>> that addresses how the hybrid interface will be represented 
>> in the OSPF and OSPFv3 MIBs. This would include the 
>> ospfIfType and ospfv3IfType.
> 
> Sure we'll add that.
> 
> Should the neighbor entry be enhanced to include per-nbr metric as well? 
> Current interface entry does include interface metrics, so it makes sense to 
> include per-nbr metric in the nbr entry.

I'd say yes. While this is not documented as a neighbor configuration parameter 
for P2MP networks, everyone has implemented it. 

Thanks,
Acee 


> 
> Thanks.
> Jeffrey
> 
>> 
>> 
>> Thanks,
>> Acee 
>> 
>> 

_______________________________________________
OSPF mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ospf

Reply via email to