Jari Arkko wrote:
Alex,
YEs, it was item 1: referenced spec (rfc2473) does have the words
about the possibility to copy the TC/Flow labels or insert
pre-configured values. However, at least one widespread
implementation doesn't implement the latter feature.
I think if some implementers realize there may be some issue in
which qos is used with mip6 then we may need to clarify 3775.
Otherwise, if in practice this aspect was not identified by some
implementers then we may continue as now until the need becomes
apparent.
I understand what you want to do. But if our situation is that
implementors are ignoring IETF RFCs, its not going to be helped by
publishing more RFCs. And including things in category 1 would be a
never-ending task. Lets focus on the other categories where we can
actually add some value. And hopefully because of you raising this
issue now here on the list some implementors are now going back to
fix their code :-)
I agree on the value adding part and I somehow disagree suggesting stuff
to implementers. I am here to suggest back stuff to specifiers.
I often notice differences between what's being specified and what's
getting widely implemented. Protocols don't make it to even a WG item
yet get deployed, and Proposed Standards protocols whose implementation
is still to happen - but I disgress.
Let me stop commenting on this and come back to Chair's reminder on
3775bis next steps.
Alex
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________
_______________________________________________
MEXT mailing list
MEXT at ietf.org
https://www.ietf.org/mailman/listinfo/mext
