Re: [Idr] New ID and request for time slot

Hi Brian,

Thx for the pointer to your draft.

It seems to me that draft-walton when combined with simply sending best 
external path (the latter some bgp implementations already support 
today) when IBGP learned path wins as the best do address all problems 
you are listing in section 1.

Could you point out what would be missing provided one would use above 
tools without additional capabilities or without need to modify bgp best 
path selection algorithm ?

Thx,
R.

> Rajiv Asati (rajiva) wrote:
>  > Hi Brian,
>  >
>  > The draft-dickson-idr-second-best-backup is not yet available for
>  > reading.
> Here's the link to the -00 version:
> http://www.ietf.org/internet-drafts/draft-dickson-idr-second-best-backup-00.txt
>  > I am certainly curious about whether and how it differs from
>  > the add-paths draft -
>  > http://tools.ietf.org/html/draft-walton-bgp-add-paths-05
>  >
>  >  
> Well, a quick review of add-paths seems to indicate several important
> differences:
> - add-paths appears limited in scope to route-reflectors, but does so
> very well
> - second-best-backup focuses more on inter-AS effects and direct
> intra-AS behavior
> 
> I expect that the two would in fact complement each other very effectively.
> 
> How exactly they would/could be merged may not be obvious just yet, but
> I don't see
> any fundamental hurdles.
> 
> Certainly, it points to lots of things there that I will likely need to
> include in subsequent versions,
> such as IANA considerations, references, etc. :-)
> 
> Thanks for pointing it out, I hadn't seen it before now.
> 
> Brian
>  > Cheers,
>  > Rajiv
>  >
>  >  
>  >> -----Original Message-----
>  >> From: idr-bounces@ietf.org [mailto:idr-bounces@ietf.org] On
>  >> Behalf Of Brian Dickson
>  >> Sent: Tuesday, February 05, 2008 6:47 PM
>  >> Cc: idr@ietf.org
>  >> Subject: [Idr] New ID and request for time slot
>  >>
>  >> I've put together a new draft, with the (ambitious!) goal of
>  >> addressing
>  >> a few of the outstanding BGP issues:
>  >> - wedgies
>  >> - persistent oscillations
>  >> - path hunting.
>  >>
>  >> I've included most of the supporting examples for the first
>  >> two, which
>  >> are clearly smaller in scope.
>  >>
>  >> I'll be doing more work (between now and IETF-71) to produce example
>  >> cases for the path hunting problem (w.r.t. the new draft).
>  >> My hope is this can reduce the level of induced churn caused by path
>  >> hunting.
>  >>
>  >> I'd respectfully request a slot in Philly to present this material.
>  >>
>  >> Comments are obviously welcome, as well as any pointers to tools for
>  >> evaluating path hunting behavior.
>  >>
>  >> Obviously, adding new attributes isn't something to be done
>  >> lightly, but
>  >> if the results are compelling enough, it may be an
>  >> appropriate path to take.
>  >>
>  >> I'd appreciate it if we can first look to see if the results are
>  >> compelling, before getting into whether they are compelling
>  >> enough. :-)
>  >>
>  >> Thanks,
>  >>
>  >> Brian Dickson
>  >>
>  >>    
>  >>> A new version of I-D,
>  >>>      
>  >> draft-dickson-idr-second-best-backup-00.txt has
>  >>    
>  >>> been successfuly submitted by Brian Dickson and posted to the IETF
>  >>> repository.
>  >>>
>  >>> Filename:        draft-dickson-idr-second-best-backup
>  >>> Revision:        00
>  >>> Title:           Enhanced BGP Capabilities for Exchanging
>  >>>      
>  >> Second-best
>  >>    
>  >>> and Back-up Paths
>  >>> Creation_date:   2008-02-05
>  >>> WG ID:           Independent Submission
>  >>> Number_of_pages: 20
>  >>>
>  >>> Abstract:
>  >>> This Internet Draft describes an enhanced way to exchange prefix
>  >>> information, to permit multiple copies of a prefix with different
>  >>> paths to be announced and withdrawn.
>  >>>
>  >>> This negotiated capability provides faster local (inter-AS) and
>  >>> global (intra-AS) convergence, reduces path-hunting, improves route-
>  >>> reflector behaviour, including eliminating both persistent
>  >>> oscillations and BGP "wedgies".
>  >>>
>  >>> Additional prefix instances have new optional transtive BGP
>  >>> attributes, to control path selection.
>  >>>
>  >>> Withdrawl of prefixes will require path attributes.
>  >>>
>  >>> Benefits are seen both when deployed intra-AS, and on inter-AS
>  >>> peering.
>  >>> Author's Note
>  >>>
>  >>> This Internet Draft is intended to result in this draft or a related
>  >>> draft(s) being placed on the Standards Track for idr.
>  >>>      
>  >> _______________________________________________
>  >> Idr mailing list
>  >> Idr@ietf.org
>  >> http://www.ietf.org/mailman/listinfo/idr
>  >>
>  >>    
> 
> _______________________________________________
> Idr mailing list
> Idr@ietf.org
> http://www.ietf.org/mailman/listinfo/idr
> 

_______________________________________________
Idr mailing list
Idr@ietf.org
http://www.ietf.org/mailman/listinfo/idr