Brian, Please see in line below.
>-----Original Message----- >From: [email protected] [mailto:[email protected]] On Behalf Of Brian Haberman >Sent: Friday, October 16, 2009 1:35 PM >To: Margaret Wasserman >Cc: [email protected] >Subject: Re: Are IPv6 auto-configured addresses transient? >I suppose there is a related question as to how an application that uses >IP addresses in referrals should behave. Would an application find it >useful to know the lifetime associated with the auto-configured addresses? I did think about referrals and now have a response. Any XYZ protocol that sends referrals in control messages should already have mechanism in place to send a control message for a referral update. If the XYZ doesn't have such control, then the XYZ protocol should add it. I would still stay away from sending lifetime of an address to an app. Likewise for sending lifetime to DNS when graceful renumbering already exists as defined in the RFCs. Personally I never get convinced for IETF taking any new work to specify any behavior if folks mention to me a corner case for a problem - like in the DNS email Mark Andrews sent. Hemant -------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
