Re: [pim] Simple join failure notification for PIM-SM multicast routing
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [pim] Simple join failure notification for PIM-SM multicast routing



Hoerdt, et al,

I did a quick review of the draft as well as reading some of Pekka's and Toerless' comments.

It seems to me that what you are calling for would best be served by a PIM-NACK message.  We've talked about doing PIM-ACK's to do reliable PIM signalling for years but most of this has centered on the use of PIM-ACK messages.

While I like the fundament concept of being able to signal tree failure, I would agree with Pekka and Toerless that we have a ways to go to put all the items in place to do this and make it effective. 

For example, let's say that there is a temporary outage or topology change that results in a Periodic Join failing to be processed. If we send this PIM-NACK message down the tree, is the last hop router supposed to send a Prune and delete the mroute state in order to tear down the now useless tree?  What if the outage has been rectified a short time later.  How does the last hop router know to "try again".

Having said the above, I don't want to discourage you from pursuing this. All I wanted to point out is that it seems to me that a new PIM control packet would be better served to accomplish this.  Ergo, the PIM-NACK that is propagated down the tree. That and (like Toerless) I'm not sure exactly how existing hosts and networks would make use of this information.

Beau Williamson
_______________________________________________
pim mailing list
pim at ietf.org
https://www1.ietf.org/mailman/listinfo/pim

Note: Messages sent to this list are the opinions of the senders and do not imply endorsement by the IETF.