[ipv6hackers] Windows 7/2008 R2 Improved Resilliency to IPv6 Floods

[Jim Small]

> > So to summarize for Microsoft KB 2750841 which retrofits Windows 8/2012
> IPv6 NCSI behavior to 7/2008R2:
> > Q1)  If you have some sort of temporary interruption in your IPv6
> connectivity at boot time will you lose IPv6 for the lifetime of the session?
> > A1)  No, from the Microsoft product manager:
> > Only a combination of no proxies, AAAA resolution success, a default route,
> and a failure to connect to the beacon - enables this functionality to activate.
> And its activation is limited to depreferencing the default route. Any scoped
> routes that have been configured are untouched.
> 
> ... which is irrelevant to the majority of users.
> 
> > Q2) Are the NCSI results cached?
> > A2) Per the Microsoft product manager, only positive results are cached.
> Negative results are re-tried at each network attach (or disable/re-enable
> interface).
> 
> ... which is also irrelevant to the majority of users (unless DHCP lease
> renewals count here).
> 
> I read and understood the feedback about what this does, and how it does
> it. My point is that for the overwhelming majority of users if there is
> a temporary interruption of IPv6 at the time they boot their systems
> they will lose IPv6 connectivity until they reboot. Of course, the
> converse is also true.
> 
> In my mind, that's a poor tradeoff. I would much rather have seen MS
> implement happy eyeballs. It would have much greater overall benefit,
> and none of the drawbacks.

I understand what you're saying - that's exactly how I felt.  I spoke to someone close to the Microsoft core networking team.  From Microsoft's vantage point the most important thing is determinism.  The problem with happy eyeballs is you have non-deterministic behavior.  For an excellent discussion of this with references, see here:
http://blog.ioshints.info/2013/03/happy-eyeballs-happiness-defined-by.html

It would be nice if there were an option to enable happy eyeballs though if the user/organization desired that behavior.  But again, I think Microsoft is afraid of the supportability/costs of a non-deterministic approach.

I' not sure I completely agree with the end result, but I understand where they are coming from.

--Jim