The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2006-Nov> msg00100



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

[mpls] probability of conifguring in a valid way "snake" testonCards in an LSR?

  • From: Kireeti Kompella <kireeti@juniper.net>
  • Date: Fri, 17 Nov 2006 07:39:46 -0800 (PST)
  • Cc: mpls@ietf.org, Esther Bar-Hall <esther.bar-hall@SeabridgeNetworks.com>

On Fri, 17 Nov 2006, Adrian Farrel wrote:

> Initial attempts to do snake tests with a box ran into the problem of not 
> being able to tell whether the snake was dead or alive until you opened the 
> box at which time, if the snake was alive, you would inevitably be bitten. 
> This drawback led to snake tests being abandoned for several years until 
> someone discovered that you could do the same test using a cage.
>
> The benefit of a cage, clearly, is that you can see in and determine from a 
> safe distance whether the snake is alive or not (except for a few cunning 
> ones that can lie particularly still). However, another problem has 
> manifested itself recently, and this may be what Esther is struggling with. 
> Unless the bars of the cage are very close together, or the snake 
> unreasonably fat, the cage often appears empty towards the end of the test.

Thank you, Adrian, it's all much clearer now!

If we used an IGP-signaled auto-mesh, would that help?  Then we can 
answer Esther's question: the topology would be full-mesh, and 
configuration (conifguration?) would be considerably simplified.

JP, you can help here.  Can you please add a 'granularity' knob for 
auto-meshes, preferably measured in millimeters?

> Adrian
>
> ----- Original Message ----- From: "Kireeti Kompella" <kireeti@juniper.net>
> To: "Esther Bar-Hall" <esther.bar-hall@SeabridgeNetworks.com>
> Cc: <mpls@ietf.org>
> Sent: Thursday, November 16, 2006 11:14 PM
> Subject: Re: [mpls] probability of conifguring in a valid way "snake" teston 
> Cards in an LSR?
>
>
>> On Mon, 13 Nov 2006, Esther Bar-Hall wrote:
>> 
>> This is truly an original message!
>> 
>> Esther, would you mind explaining what this is about?
>> 
>>> -----Original Message-----
>>> From: Kireeti Kompella [mailto:kireeti@juniper.net]
>>> Sent: Wednesday, November 08, 2006 21:56
>>> To: LE ROUX Jean-Louis RD-CORE-LAN
>>> Hi,
>>> 
>>> I need to know whether there is a way to fully populate a Cage with
>>> cards and run a "snake" test on all ports with two ports of TE
>>> connected to the cage, where the NE functions as an LER/LSR?
>>> 
>>> Please advise on the matter and how will the topology will be and the
>>> configurations?
>>> 
>>> Any help will be appreciated.
>>> 
>>> 
>>> 
>>> _______________________________________________
>>> mpls mailing list
>>> mpls@lists.ietf.org
>>> https://www1.ietf.org/mailman/listinfo/mpls
>>> 
>>> 
>>> 
>>> 
>>> _______________________________________________
>>> mpls mailing list
>>> mpls@lists.ietf.org
>>> https://www1.ietf.org/mailman/listinfo/mpls
>>> 
>> 
>> Kireeti.
>> -------
>> 
>> _______________________________________________
>> mpls mailing list
>> mpls@lists.ietf.org
>> https://www1.ietf.org/mailman/listinfo/mpls
>> 
>> 
>> 
>

Kireeti.
-------

_______________________________________________
mpls mailing list
mpls@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/mpls