Re: [BEHAVE] proprietary implementation v.s standardisedprotocols//re: draft-xu-behave-nat-state-sync-00

[Andrew Sullivan <[e-mail]>]

On Tue, Dec 01, 2009 at 03:07:11PM -0500, Simon Perreault wrote:

> The point was that you cannot fail over clients from one Pref64::/n to another
> because applications will remember addresses. This is true, to some extent. If
> it was 100% true then you wouldn't be able to change anything once you publish a
> DNS record. In reality, hosts are rebooted, applications are restarted, and web
> pages are reloaded. So you have some wiggle room.

"Reloading a web page" often doesn't help.  I have experience of at
least one popular browser that reaches down and remembers the A record
for the target host.  Then the magic DNS load balancer decides to pull
that IP out of the mix, and stuff starts breaking randomly.  (One of
the many ways in which DNS load balancers don't work the way their
deployers seem to think they do.)   That said,

> partial fix is better than just shutting of the Internet to your
> customers. If it means they have to close Outlook and restart it, so
> be it.

I completely agree here.

A

-- 
Andrew Sullivan
[e-mail]
Shinkuro, Inc.