Re: [dhcwg] comments on draft-ietf-dhc-lifetime-01.txt

[Stig Venaas <[e-mail]>]

On Thu, Aug 12, 2004 at 08:31:01AM -0400, Joe Quanaim wrote:
> Stig Venaas wrote:
[...]
> > I think stateless only might be too strict.
> >
> > One example I have in mind is:
> >
> > If a client sends a request message to obtain addresses and other config
> > options and it does not get any addresses but receives other config
> > data. Then the client may use those options and not make a
> > Request/Information-Request, right? You say client needs to do something
> > when it no longer has addresses, but if you have IPv4 addresses, and the
> > only IPv6 DHCP server says it has no addresses, then I think it's ok for
> > the client to just use what it got without switching to stateless mode,
> > and at some later point make a new Request, still not using stateless
> > mode.
> 
> I am not sure that the lifetime option is necessary when using stateful.  I 
> would consider this example stateless because the client has not successfully 
> negotiated any (IPv6) addresses.  Here I think you are right, though, a 
> client should use the lifetime option.  If it has any addresses, I think it 
> should ignore the lifetime option.

OK, I can agree with you if you define stateless that way. I'm thinking
of stateless as defined in RFC 3736 using only Information-Request/Reply.
My point is that it might be useful to use it with Request/Reply as well.

So, my suggested wording should be ok then I understand you correctly.

> > How about saying this instead?:
> >
> >    The lifetime option is mainly intended for Stateless DHCP service as
> >    specified in RFC 3736.  If the client receives IA Address options
> >    containing lifetimes, the lifetime option should be ignored.  The
> >    client should get updated configuration data from the server when it
> >    renews the addresses.
> >
> > Another thing I'm wondering about is if T1, T2 are zero or infinity. I
> > think the lifetime option might make sense then. In 3315 it says:
> >
> >    If T1 or T2 is set to 0 by the server (for an IA_NA) or there are no
> >    T1 or T2 times (for an IA_TA), the client may send a Renew or Rebind
> >    message, respectively, at the client's discretion.
> >
> > If you get e.g. 0, it might be good to have some sort of lifetime for
> > the other data, and I guess you might then do renew/rebind when the
> > lifetime expires.
> >
> 
> Even if T1 and T2 are zero or infinity, you have the lifetime of the address 
> to fall back on.  RFC 3315 says:
> 
>    Note that an IA_NA has no explicit "lifetime" or "lease length" of
>    its own.  When the valid lifetimes of all of the addresses in an
>    IA_NA have expired, the IA_NA can be considered as having expired.
> 
> And says something similar for IA_TA's.  There is still an issue when the 
> lifetime of the address is set to infinity, but this seems like a special 
> case and one that might be used intentionally for one time configuration.

Right, sounds reasonable.

Thanks,

Stig

_______________________________________________
dhcwg mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/dhcwg