Re: [rrg] IPv4 & IPv6 routing scaling problems

[[e-mail]]

 
In einer eMail vom 14.02.2010 07:46:03 Westeuropäische Normalzeit schreibt  
[e-mail]:

On Sat,  Feb 13, 2010 at 5:03 AM,  <[e-mail]> wrote:

>  loop: from PG1 to PG2 back to PG1. Today I see opposition against TARA
>  because the network inside a geopatch might partition. Yes, this may  
happen.
> But dealing with partitions starts with getting from one  partition to the
> other. And I can only offer a loop: out to some  neighbor geopatch, from
> there back to the other partition of the own  geopatch.

if all you know is PG1, no subnets/sublocators/identifiers,  how does
pg2 know not to send traffic back the same link it came  from?

(note this all seems pretty far out of rrg  though)



So let me answer your question just in the context of a Topology  
Aggregating Routing Architecture:
Each TARA-router would realize that its geopatch is partitioned by  
receiving BGP-UPDATE's containing TARA-links of some upper zoom into the own  
geopatch which it wasn't able to produce itself.
Loopfree, minimally detouring routes via neighboring geopatches can be  
determined and can be enforced for being taken.
 
Heiner