kre,
How about this (changes in the second sentence):
"The method by which the flow state is cleared from the IPv6 nodes is to be defined by
the flow state establishment method used to set up the state. This implies that IPv6
nodes unable to classify a packet to an existing flow SHOULD NOT establish any
flow-specific state unless so instructed by a flow state establishment method. With
some flow state establishment methods, signaling new state simultaneously clearing the
old state from the new path may be sufficient. A mechanism with a sufficiently long
timeout period before reusing the Flow Label values can also be used."
I also fixed the grammar nit you pointed out, thanks.
Jarno
> -----Original Message-----
> From: ext Robert Elz [mailto:[EMAIL PROTECTED]]
> Sent: 5. elokuuta 2002 16:20
> To: Rajahalme Jarno (NRC/Helsinki)
> Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]
> Subject: Re: Flow label draft issues & new text
>
>
> Date: Mon, 5 Aug 2002 09:53:49 +0300
> From: [EMAIL PROTECTED]
> Message-ID:
> <[EMAIL PROTECTED]>
>
> | any suggestions on how to make the text clearer about this?
>
> [EMAIL PROTECTED] said:
> | Send text.
>
> Me and my big keyboard...
>
> Well..
>
> The method by which the flow state is cleared from the
> IPv6 nodes is
> to be defined by the flow state establishment method used to set up
> the state.
>
> That's fine.
>
> This implies that IPv6 nodes SHOULD NOT establish any
> flow-specific state unless so instructed by a specific flow state
> establishment method.
>
> That's fine too, provided that you assume that the reader has actually
> read the definitions, and noticed:
>
> Flow state A control mechanism used to set up the flow
> establishment method state. A flow state establishment method can
> be either
> - Dynamic, under source or destination node
> control (e.g. RSVP),
> - Quasi-dynamic, under network management
> control,
> - Static, through manual configuration, or
> - Algorithmic (e.g. load spreading)
>
> (grammar nit: I don't think "either" is correct to describe a choice
> among more than 2 things, make it "any of").
>
> The problem is that many simply assume that they know what
> all the words
> mean, and treat the definitions section, like acknowledgments, and
> security considerations, as something of no relevance, and skip it.
>
> So, perhaps add
>
> In particular, nodes should not simply react to observing
> the flow labels of traffic passing through. This is not
> a flow state establishment method, and, in particular, gives no
> hint as to when packets may now be part of a different flow than
> previous packets similarly labeled.
>
> The rest ...
>
> With some flow state establishment methods,
> signaling new state simultaneously clearing the old state from the
> new path may be sufficient. A mechanism with a sufficiently long
> timeout period before reusing the Flow Label values can
> also be used.
>
> is also OK, that's just hints to flow state setup types (ie: working
> groups and the link) as to what they might like to define in
> their protocols.
>
> kre
>
> --------------------------------------------------------------------
> 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]
> --------------------------------------------------------------------
>
--------------------------------------------------------------------
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]
--------------------------------------------------------------------
