RE: Call for v6ops agenda items

[<teemu.savolainen@nokia.com>]

Hi,

Thank you very much for comments. I'm preparing an update. Comments inline:

> * there already is a well working protocol for PD: DHCPv6; although I
> like the idea of finally getting rid of it

Right. IMHO as is proven by 6RD, there's interest to have prefix delegation also without DHCPv6 if easily doable.

> * the unique-bits part is too large to work on more than one level, so
> it
> would be reserved to the ISP-to-User-line, but then again SLAAC is also
> reserved to the link level of routing

I don't agree, the draft intents to say unique-bits can be of about any length. For example:
- if length of unique bits is 1, there can be two routers below the gateway, which both would get delegated prefix of: [SPD-prefix][0/1][subnet ID]. Now if the SPD prefix is say 40 bits, and V6UNIQUE is 1 bit, the router would get /41, right? Then this router could further choose to delegate say /56 by configuring downlink routers by provisioning them with: [SPD-prefix+1][V6UNIQUE of 14 bits of length][subnet ID]. Right?

In fact, this unique bits could be minimum of 1 bit and maximum is many bits (I have to think about that), right?

> * semantically it's not the right protocol for this: SLAAC is
> transmitted
> on the same link it configures, PD is transmitted on the upstream link
> of
> those (different links) that are configured

With SLAAC you refer to the case where bits are taken from the /64 prefix configured with SLAAC? Right, but what is the significant difference to 6RD that uses uplinks IPv4 address?

> * as a router admin I would never accept PD statelessly - i.e. without
> having asked for it; this might be just too much paranoia on my part
> though

I'm not sure I see the difference, in both cases network has to be trusted... Also, you can trust to the context: e.g. having a dedicated 3GPP APN for this purpose. 

> * stateless also means that the router has to reserve enough space for
> EVERY client, because it cannot know whether the client wants to use
> it;
> with a stateful protocol it only needs to reserve space and route the
> prefix once the client asks for it

Right, but this depends on the context as well. If you limit stateless delegation only to parts of the network where you know all nodes *are* routers, then the waste is mitigated. Only if the routers' needs vary a lot (some require /48 while most would settle for /60), but all are allocate /48, there would be waste. In 3GPP e.g. this could be mitigated with an APN dedicated for router use - all UEs using this APN would be once (potentially) needing prefixes. 

This is the essential tradeoff in the stateless approach - there's no free lunch.. But which one is more expensive, IPv6 address space, or fully dynamic system?

> * for most OSes it has to be implemented in a user-space program
> anyway,
> since the kernel cannot know which interfaces to configure, but in
> userspace it does not matter whether I implement DHCP rapid commit or
> "stateless" PD with a request message - the complexity is the same

True. On the requesting router side the savings are indeed less - the main benefits would be on the network side.

> I also have a problem with some of the uniqueness-source bits:
> 
> P: should be restricted to prefixes announced in the same SLAAC message

Same as what? The service provider prefix would not be delivered in SLAAC. Anyhow, there's issue if there are multiple /64s in RA (while in 3GPP there currently are always one). Updating draft..

> I/2: should be unified into just I, since it is guaranteed to exist and
> is
> dependent on layer 2 IDs anyway

Done, but added "T": tunnel identifier. What I'm thinking here is that e.g. the GTP TEID of 3GPP access could be unique part. Each bearer has this locally unique 32bit identifier. I think this requires further clarification.

> 4: I have a really big problem with this: it mixes protocols and
> requires
> a lot of dual-stack logic to exist on the host - you can make it a MAY,
> but then DHCPv6-PD has to exist as well as a backup

What do you think about 6RD?

> D: it shouldn't mention DHCP directly, but instead use the word
> "stateful"
> as other RFCs do, but then again - how to know which IP is meant? The
> interface can have multiple IPs all configured through different
> stateful
> protocols and an independent process will not know through which
> protocol
> they came. BTW: why would one use stateless PD if there is stateful
> address assignment anyway?

Right.. This was here for sake of "completeness".. but I'll throw it away.

> If multiple bits are set - is that legal? What should the host do?

Updated to say it is not legal.

> I also think the ping test should be a MUST, not a MAY.

Updated.

Thank you,

Teemu