But, in these cases both addresses are stateless so using how the address was assigned would still not resolve the issue.
So, that is a related, but separate issue? Probably the only solution to that is to have some stickiness factor where a host continues to favor that address for a period of time. An easy solution may simple by to divide the perferred lifetime by some factor and compare those. That way, small changes in the preferred lifetime don't alter the result. - Bernie -----Original Message----- From: Stig Venaas [mailto:[EMAIL PROTECTED] Sent: Thursday, October 26, 2006 4:48 PM To: Bernie Volz (volz) Cc: ipv6@ietf.org Subject: Re: address selection and DHCPv6 Bernie Volz (volz) wrote: > I would think that how an address is assigned shouldn't enter into this. > I can't see that it matters. > > What really matters is the lifetimes associated with the address. The > longest lifetime address is probably the best to use since it is the > most stable. [Ignoring privacy and other related issues.] That's sort of what I thought at first as well. But if you use remaining lifetime (and not the initial) you might easily end up in situations where you fluctuate between several addresses. Not so good. As an example, imagine two routers regularly sending RAs for one prefix each, then you might end up always using the prefix you last got an RA for, possibly changing every few seconds depending on the adv. rate. Stig > > Perhaps this gets at what you want anyway, since manually assigned > addresses would presumable have the longest lifetime? > > - Bernie > > -----Original Message----- > From: Manfredi, Albert E [mailto:[EMAIL PROTECTED] > Sent: Thursday, October 26, 2006 3:06 PM > To: Durand, Alain > Cc: ipv6@ietf.org > Subject: RE: address selection and DHCPv6 > > Except that from what others have said, that might not be the desired > goal. Perhaps for privacy or other reasons, a most stable address choice > might not be optimal. > > I originally thought that would be the best choice, but ... > > Bert > > > >>-----Original Message----- >>From: Durand, Alain [mailto:[EMAIL PROTECTED] >>Sent: Thursday, October 26, 2006 2:00 PM >>To: Bernie Volz (volz); James Carlson; Vlad Yasevich >>Cc: ipv6@ietf.org >>Subject: RE: address selection and DHCPv6 >> >>The question is not to get an absolutely stable address, >>but to make sure that in case multiple addresses are defined, >>the one with the highest likelyhood of stability is selected. > > > -------------------------------------------------------------------- > IETF IPv6 working group mailing list > ipv6@ietf.org > Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 > -------------------------------------------------------------------- > > -------------------------------------------------------------------- > IETF IPv6 working group mailing list > ipv6@ietf.org > Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 > -------------------------------------------------------------------- -------------------------------------------------------------------- IETF IPv6 working group mailing list ipv6@ietf.org Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
