On 08  May 2011, at 19:47 , Brian E Carpenter wrote:
> But it's really playing with words to assert that a firewall which
> chooses to overwrite the field is "supporting" it in the sense
> intended by the phrase "Hosts or routers that do not support the
> functions of the Flow Label field...".

It is not playing with words.  Those security gateways normally
are also routers, using the formal IPv6 definition of a router, 
often participate in routing protocols (e.g. eBGP in the case
of multi-homed uplinks from an end-site; OSPF or RIP within an
end -site) as routers, and often use the Flow Label value as input 
to their forwarding decisions.

> The technical issue is that load balancing based on flow label values
> won't work properly if middleboxes (or attackers) change the values
> arbitrarily.

As near as I can tell, everyone here agrees on the objective of
enabling the Flow Label to be useful for load-balancing.

As my note of a few moments ago suggested, there is a way to address
both the operational security concerns and the load-balancing objectives.
Rather than repeat myself, I'll just refer back to that note.


> Thomas has argued that if the label MUST NOT be changed, we should
> remove any suggestion of cases in which it's OK to change it. Followed
> to its logical conclusion, that means it's not OK for a firewall to
> change it.

The crux of his concern seems to be that the current text tries 
to have things both ways, which ambiguity is long-standing, 
as I noted in my weekend email.  

I hope, perhaps overly optimistically, that the formula I proposed
in my note of a few moments ago will find middle ground that most
folks find acceptable.

Cheers,

Ran

--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to