Alex, Thanks for pointing this out. What do you think about the following as the replacement text?
============= The rate-limiting parameters SHOULD be configurable. In the case of a token-bucket implementation, the best defaults depend on the computational power of the device and the speed/bandwidth of the interfaces that it has (e.g., a high-end router vs. an embedded host, a 10G ethernet vs. a T1 interface). For example, in a small/mid -sized device with limited bandwidth interfaces, the possible defaults could be B=10, N=10/s. ============== Regards Mukesh > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of > ext Alex Conta > Sent: Thursday, August 05, 2004 11:17 AM > To: IPv6 > Subject: Section 2.4, item (f) of draft-ietf-ipngwg-icmp-v3-04.txt > > > The last paragraph of Section 2.4, item (f) of > draft-ietf-ipngwg-icmp-v3-04.txt is pointing to ICMP rate limiting > configurability on a per node basis. > > ICMP Rate-limiting configurability makes sense also on a per > interface > basis. > > In a case of a router having different speed/bandwidth > interfaces, for > example IPv6 interfaces on fast links, as well as slower links, the > adequate ICMP rate-limiting parameters on a 1G or 10G > Ethernet, could be > quite different from a T1, or T3 link. > > Regards, > Alex > -------------------------------------------------------------------- IETF IPv6 working group mailing list [EMAIL PROTECTED] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------