No, I think the immutability property must be specified in general. You can't tell by looking at an opaque 20 bit number whether it's being used for one purpose or another, so without immutability being required, it is a useless field (at least for intserv or diffserv).
I agree that in the end we need very few words. The sentence you quote, plus In all cases, the flow label in the packet delivered to the destination node must be identical to the flow label in the packet sent by the source node. would do it for me. All the rest of the text is rationale. Brian Craig Dunk wrote: > > I like the rule below as it seems to nicely separate responsibilities. > > Will it be self evident when (others) are applying the specific mechanisms > mentioned that the flow label (being unprotected) should not be used in a > way that circumvents normal routing rules or security policies? > > Craig. > > -----Original Message----- > From: Margaret Wasserman [mailto:[EMAIL PROTECTED]] > Sent: December 20, 2001 11:50 AM > To: [EMAIL PROTECTED] > Subject: Re: draft-rajahalme-ipv6-flow-label-00.txt > > Hi All, > > I am wondering why we should specify the use of the flow label as > part of the base IPv6 specifications at all. Why do we need > these rules as part of IPv6? > > The flow label does not, by itself, provide any useful information > that a router can use to classify a flow and/or optimize packet > handling. In fact, without knowledge of a specific signalling > mechanism or flow-establishment mechanism, the router can't > use the flow label for anything at all. For example, a router > cannot use a flow label for QoS queuing or to optimize hop-by-hop > header processing, unless the router is aware of the signalling/ > flow-establishment mechanism in use, since it will not > know the flow lifetime. > > Any rules governing how the flow label is chosen, modified, > authenticated and/or interpreted will be specific to the > signalling/flow-establishment mechanism in use. And, all of > the nodes/routers that are utilizing one of these mechanisms > will be aware of the mechanism. > > So why not specify the semantics/use/etc. of the flow label as > part of the signalling/flow establishment protocols? > > The IPv6 specifications could merely include the current rule: > > "Hosts or routers that do not support the functions of the Flow Label > field are required to set the field to zero when originating a packet, > pass the field on unchanged when forwarding a packet, and ignore the > field when receiving a packet." [from RFC 2460]. > > This rule should properly protect the flow label, so that > signalling/flow-establishment mechanisms can use the flow > label, as needed by the specific mechanism. > > Margaret > -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
