> > I am just curious if all other TLVs/subTLVs of BGP's UPDATEs must be > present with value NULL (0) to emphasize that it is not accidently omitted? >
They do not need to be present with NULL (0) value, but no one is drawing any conclusion on missing attributes. Here you still want to implicitly take next hops and treat it as tunnel endpoint - as John mentioned that could be possible if spec would define such behavior when tunnel attribute is present and Remote Endpoint Sub-TLV is not there. But I am not sure however if there is sufficient value to it. [Linda] Isn’t the Source Address of the UPDATE message same as the > originating node of the UPDATE message? > No. Think about going via RRs as a basic example. Thx, R. >
_______________________________________________ rtgwg mailing list [email protected] https://www.ietf.org/mailman/listinfo/rtgwg
