Re: [v6ops] PI [ULA draft revision #2 Regarding isolated networks]

[Mark Andrews <[e-mail]>]

In message <[e-mail]>, Ted Lemon writes
:
> On Jun 1, 2014, at 10:28 PM, joel jaeggli <[e-mail]> wrote:
> > So don't asssume that every application needs to be aware of and test
> > for which path is working, not only is that needless complexity pushed
> > up into the stack. let your routers do their job and cease routing when
> > their upstream connectivity is lacking.
> 
> Right, but this doesn't actually work.

Setting the prefix's preferred lifetime to 0 in RAs when the uplink
goes and restoring it when the uplink works does work for non-broken
stacks.  Yes this is crude but so is advertising default.  Withdrawing
default is what routers should be doing when there isn't a working
default. 

There is also nothing to prevent a stack trying multiple source
addresses when it hasn't been explicitly bound to a source address
by the application.  The stack doesn't have to complete TCP handshakes
if it doesn't need to.

If all else fails the application layer is free to explictly try
different source addresses.  Hiding this detail in a library routine
is not hard.

> _______________________________________________
> v6ops mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/v6ops
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: [phone]                 INTERNET: [e-mail]