Re: [dhcwg] Next step(s) for draft-ietf-dhc-stable-privacy-addresses -> abandon work? / IA_NA applicability

["Bernie Volz (volz)" <[e-mail]>]

> So the clear benefit is that you can have multiple DHCPv6 servers operating on the same network, and you get address stability without an additional protocol.

This concept is flawed if the servers aren't cooperating. If there are multiple servers that don't share lease data, how does server 1 know whether the lease was leased to another client or renewed, if server 2 ends up communicating with the client? Server 1 may also expire the lease (and remove FQDN or other data). Or when a leasequery is done, the servers will return conflicting data (one says not leased, other says leased).

I just don't buy that is has any real benefit.

But I am going to refrain from discussing this topic any further as my position (not as co-chair, but as an individual) is clear. Once we get different people to weight in (either on the mailing list or at IETF-93) to determine consensus we can figure out the appropriate next step.

- Bernie

-----Original Message-----
From: Fernando Gont [mailto:[e-mail]] 
Sent: Thursday, April 09, 2015 1:25 PM
To: Lorenzo Colitti
Cc: [e-mail]; Bernie Volz (volz); Ted Lemon
Subject: Re: [dhcwg] Next step(s) for draft-ietf-dhc-stable-privacy-addresses -> abandon work? / IA_NA applicability

On 04/08/2015 09:18 PM, Lorenzo Colitti wrote:
>     For the time being, I think we should not consider the 100% stateless
>     version of this. After all, the man goal of this document is an address
>     selection scheme for DHCPv6, rather than an improvement in terms of "now
>     DHCPv6 can be fully stateless".
> 
> If this algorithm does not allow stateless operation, and still 
> requires keeping (and synchronizing) a lease database, then I really 
> don't see the point of specifying this algorithm at all, as it 
> provides no advantages over stateful random assignment.

This algorithm doesn't require keeping a lease database. DHCPv6 does.

And no, as noted in my other email, it does not require synchronization of the lease database. IN the case of failover, rather than "renewing"
or "rebinding" the same address, the node will be leased the same address as if it was a brandnew address -- same end result.

So the clear benefit is that you can have multiple DHCPv6 servers operating on the same network, and you get address stabiity without an additional protocol.

--
Fernando Gont
SI6 Networks
e-mail: [e-mail]
PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492