RE: 64bit MAC addresses and SLAAC

["Pascal Thubert (pthubert)" <[e-mail]>]

> > >
> > Is this really, deep down, the nature of SLAAC? Or the consequence of
> > not having a very fast and reliable DAD? Arguably one could autoconfig
> > even IPv4 within a /24 if there's a broker that serializes the
> > requests and rejects the duplicates.
> 
> We have that broker, it is called DHCPv6.

As of today? Depending on what you expect from that broker, e.g., who gets to select the IID, can you use the broker for Address Resolution, etc... 
DHCP is a swiss knife. With a few more blades it would fit the bill nicely. ND could too. There are other candidates but I'd rather upgrade those two to start with.

> 
> Absent such a centralized service, a host would have a hard time proving that a
> address unique, for example, there might be a network split.

We agree

> Current DAD is certainly not good enough if there are regular address collisions

Define "regular". I will agree if it is the opposite of quasi-impossible, which is how we get SLAAC to work.
Thus your original shortcut that SLAAC=>randomIID though deep down it does not, it's just how we do it with distributed DAD.

> and I doubt anybody is willing to come up with a better DAD just to have nicer
> looking addresses in SLAAC.

It's not just DAD, it's the whole ND system that has aged and needs a lifting. Think wireless, overlays, NBMA meshes, large L2 domains...

Keep safe;

Pascal