Open Stack

On Thu, Feb 12, 2015 at 6:36 AM, Salvatore Orlando <sorlando at nicira.com> wrote:
> - I agree with Carl that the IPAM driver should not have explicit code paths
> for autoaddress subnets, such as DHCPv6 stateless ones. In that case, the
> consumer of the driver will generate the address and then to the IPAM driver
> that would just be allocation of a specific address. However, I have the
> impression the driver still needs to be aware of whether the subnet has an
> automatic address mode or not - since in this case 'any' address allocation
> won't be possible. There already comments about this in the review [1]

I had another thought on this.  You say that "any address allocation
won't be possible".  I wonder if this is really true.  The EUI-64
space is only one part in 64k of total /64 space.  IPAM could allocate
from the rest.  There could be use cases for mixed modes in the
future.

I think best practice for any IPAM should be to steer clear of
allocating address which could be EUI-64 addresses.  Even if it
ignores it completely, what are the chances that IPAM will allocate
from that sub-space *and* allocate an address that would conflict?
I'd say pretty low.  If it did happened, IPAM would reject the address
computed and the port create would fail I think.  You'd try again and
it would likely succeed.  I may be over-simplifying.

Carl