Some suggestions: TTL: - should refer to “TTL or hop count” (IPv6 calls this hop count because it no longer intends to reflect time, as was originally intended for IPv4) - upon encapsulation, the outer header TTL/hopcount should be set by policy (not technology), i.e., “as desired” - upon decapsulation, the inner header TTL/hopcount should be modified by policy as well. it might be useful to suggest that the inner one never be incremented and that it might be decremented to reflect the cost of tunnel forwarding, but it need not be modified at all Fragmentation: I’ve never seen “COULD” used this way and it’s incorrect. They ARE described in that draft. It would be more correct, IMO, to say: The specific requirements for fragmentation and tunnel configuration SHOULD follow the recommendations in [I-D.ietf-intarea-tunnels]. Joe |
