inline

On Tue, Mar 29, 2011 at 5:45 PM, Greg Mirsky <[email protected]> wrote:
> Dear Authors,
> please kindly consider my comments and questions below (though some could be
> duplicates of already received):
>
> section 4.3 Fast Flooding has reliability as first requirement. I've noticed
> that Section 4 Operation doesn't list Auto-discovery or Registration as part
> of Fast Failure Notification mechanism. Is it assumed that an IGP will build
> the list of addressees, participants of the Fast Notification?
> how would node(s) that detected a failure would achieve reliable transport
> of notifications?

This is through capability advertisement. It was in the -00 but was
erroneously omitted in -01. Will be re-added.

> if full set of destinations is not known what indicates that notification
> reached all participants?
> when requiring reliability of notification transport you consider
> "notification received" or "notification processed"

There is no ack for FN. Redundancy is achieved by using redundant
trees. OSPF Flooding ensures consistency in the worst case.

> section 5.2 I believe that in the scenario you present not only node B will
> flood notifications, but node A will flood them as well. That would not
> change convergence time if only to improve it in some scenarios but it might
> be worth to mention that notifications will be generated by both ends of a
> failed link.

That depends on the failure and its detection. It will only improve
things and has been mentioned in the related drafts.

>
> Regards,
> Greg
>
> On Tue, Mar 29, 2011 at 8:49 AM, Sriganesh Kini
> <[email protected]> wrote:
>>
>> Just an FYI to the list
>>
>> This draft
>> http://tools.ietf.org/html/draft-kini-ospf-fast-notification-01 was
>> presented at the OSPF WG mtg today.
>>
>> Thanks for the comments/questions at the mic. We will submit a new version
>> addressing the comments.
>>
>> Note that the other drafts related to Fast Notification (FN) are
>> draft-lu-fn-transport and draft-lu-fast-notification-framework. These were
>> presented in RTGWG.
>>
>> Thanks
>>
>> - Sri
>>
>> _______________________________________________
>> OSPF mailing list
>> [email protected]
>> https://www.ietf.org/mailman/listinfo/ospf
>>
>
>
> _______________________________________________
> OSPF mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/ospf
>
>



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

Reply via email to