[rbridge] Sec 2.3 in draft-touch-trill-rbridge-prob-00
Guillermo Ibáñez
gibanez at it.uc3m.es
Wed Nov 9 12:14:25 PST 2005
Alia Atlas wrote:
>"It would be more useful for subnet configuration to be tolerant of
> such transients, e.g., supporting alternate, backup paths.
>
> [QUESTION: is there more to say here?]
>
> Contrast this to network layer intradomain and interdomain routing,
> both of which include provisions for backup paths. These backups
> allow routing to be more stable in the presence of transients, as
> well as to recover more "rapidly when the transient disappears. "
>
>I found the description here to be rather confusing. Are you saying
>that the possibility for the use of additional links permits easier
>routing convergence? Or are you suggesting that
>alternate next-hops can be pre-computed and used on a failure - as is
>described in the IP fast-reroute work in rtgwg? I'd assume the
>former. If that's the case, then I think the language needs some
>clarification.
>
>
May be the terminology is confusing, but it is a must to obtain fast
reconfiguration as required by rbridges.
(I will be again advocate of RSTP as example, that includes alternate
port role, able to be root port via different path to root bridge, for
faster reconfiguration.
>Alia
>_______________________________________________
>rbridge mailing list
>rbridge at postel.org
>http://www.postel.org/mailman/listinfo/rbridge
>
>
>
More information about the rbridge
mailing list