cloudstack-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Daan Hoogland <daan.hoogl...@gmail.com>
Subject Re: IPv6 in VPC (was Re: IPv6 plan - questions)
Date Fri, 17 Jan 2014 14:28:45 GMT
Ok, I though those could come from the same vpc range.

On Fri, Jan 17, 2014 at 2:58 PM, Marcus Sorensen <shadowsor@gmail.com> wrote:
> From what I understand, SLAAC only works with /64s, larger breaks
> various discovery protocols and is against RFC. Half of the address is
> the prefix and the other half is (mostly) MAC. What you're describing
> would work if we didn't want to do SLAAC, but would require an
> alternate means of assignment.  This sort of goes back to me wanting
> to be able to assign multiple ranges to a network, say a "SLAAC /64"
> and a "Manually assigned /64" by providing a field to tag the prefix
> with a type.
>
> On Fri, Jan 17, 2014 at 6:43 AM, Daan Hoogland <daan.hoogland@gmail.com> wrote:
>> On Fri, Jan 17, 2014 at 8:45 AM, Marcus Sorensen <shadowsor@gmail.com> wrote:
>>> guest networks, my initial preference would be for SLAAC, but I think
>>> ultimately we'd want to be able to assign multiple ips to a guest.
>>> With the 64 bits of the SLAAC space dedicated to all of the unique MAC
>>> address possibilities, we can't really do that. We may want to
>>> consider DHCP with static assignments from the beginning.
>>
>>
>> Wouldn't an admin that requires this just have to assign bigger
>> networks? /60 for the tiers and /56 for the vpc... Seems like this is
>> not an argument against SLAAC in favor of DHCP. do tell me I am
>> stoned, Daan

Mime
View raw message