The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2005-Mar> msg00026



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

[mpls] COMMENTS/CONCERNS/... aboutdraft-defeng-mpls-mini-fast-rerouting-00.txt

  • From: Alia Atlas <aatlas@avici.com>
  • Date: Tue, 08 Mar 2005 12:28:35 -0500
  • Cc: mpls@ietf.org, liujongyfw@chinamobile.com, l.b@huawei.com, duanxiaodong@chinamobile.com, sunshaoling@chinamobile.com
  • X-Avici-MailScanner-Information: Please contact the ISP for more information

The method in the document seems to compute a "second-best" path in the 
event of the isolated local failure.  Unfortunately, the second-best path 
may not be one that can actually taken in the network successfully until 
the network has converged.   This is one of the complications with IPFRR.

JP's modified example below demonstrates this problem.

Alia

At 12:10 PM 3/8/2005, JP Vasseur wrote:
>3) More importantly ... your mini-FRR scheme works well on a mini-topology 
>....
>
>In your example, please try the two following changes:
>
>-> make the shortest path from R3 to R5 be: R3-R1-R2-R5
>-> insert a router between R2 and R3 so that R3 cannot detect the R2's 
>failure.



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