Vijay,

I think it's been safe to say for many years that *all*
IP addresses, of either family, are transient, with limited
exceptions (i.e. stable servers). I don't see that IPv6
autoconf addresses are qualitatively different in that respect.

It is a deep design bug that when applications obtain an address
from lower layers, it doesn't come with a lifetime attached.
Addresses derived from A or AAAA records, DHCP or DHCPv6, etc.,
all have finite lifetimes, and it's very sad that applications
don't know it.

Actually let me plug the GROBJ BOF in Hiroshima, and the associated
draft-carpenter-behave-referral-object (version 01 coming soon).
Address lifetime becomes an especially serious problem when
referring addresses from one host to another.

   Brian

On 2009-10-08 05:25, Vijayrajan ranganathan wrote:
> Hi Everyone,
> Is there a notion that auto-configured IPv6 addresses based on
> globally unique prefixes are transient compared to manually configured
> ones?
> I know that we could configure them to have infinite lifetimes & such,
> but I am thinking of a large IPv6 deployment where these addresses are
> expected to be up 24/7. I mean, just the fact that lifetimes could be
> modified on the router, the advertising router could become
> unreachable etc.
> 
> As an IPv6 application developer, would I have to factor in this
> "transiency" of autoconf addresses in my design all the time? How safe
> & normal
> is it to replace all manual IPv6 address configuration with
> auto-configuration in a large IPv6 deployment esp in an environment
> that is very sensitive
> to non-availability of addresses?
> 
> Another related question, is it common for a site's global prefix(es)
> to change? In this regard, are they any different from an IANA
> assigned IPv4
> network-id for example?
> 
> Thanks & Regards
> Vijay
> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> [email protected]
> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> --------------------------------------------------------------------
> 
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to