On Thu, Nov 10, 2016 at 04:19:14PM +0100, Gert Doering wrote: > On Wed, Nov 09, 2016 at 06:59:53PM +0000, Sriram, Kotikalapudi (Fed) wrote: > > The data plane would perform the usual uRPF check: Does the SA in the data > > packet > > belong in a prefix in the RPF list for the interface it was received on? > > This, actually, is not "the usual uRPF check". > > Having implementations that could tack arbitrary "RPF lists" to an > interface would be very nice, but this is more like "auto-generate ACLs > based on prefix info" than "RPF" which stands for "reverse path filter" > (not sure about the "filter" bit, though)
This summarizes my hallway feedback to Sriram. As noted during mic chat, the existing RPF checks take advantage of existing line card implementations to do their thing. The general necessary feature would be to extend the line cards to take arbitrary lists that may have nothing to do with the existing FIB contents. -- Jeff _______________________________________________ GROW mailing list [email protected] https://www.ietf.org/mailman/listinfo/grow
