The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2003-Jul> msg00060



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

questions about draft-ietf-mpls-rsvp-lsp-fastreroute-02.txt

  • From: Terry Lee <terrylee@huawei.com>
  • Date: Tue, 15 Jul 2003 16:46:14 +0800
  • Cc: "'Prabakaran T Sampath'" <prabakarts@future.futsoft.com>, "'mpls@uu.net'" <mpls@UU.NET>
  • Importance: Normal

Title: 邮件
Hi,
Please find my comments in paragraphs [Terry Lee].
 
-----Original Message-----
From: Alia Atlas [mailto:aatlas@avici.com]
Sent: Tuesday, July 15, 2003 4:38 PM
To: Terry Lee
Cc: Prabakaran T Sampath; 'mpls@uu.net'
Subject: RE: questions about draft-ietf-mpls-rsvp-lsp-fastreroute-02.txt

At 09:25 PM 7/14/2003, Terry Lee wrote:
Thanks a lot.
 
I have another question. In section 6.4, it states as following :
 
   6.4 Signaling for Facility Protection
 
      A PLR may use one or more bypass tunnels to protect against the
      failure of a link and/or a node.  These bypass tunnels may be
      setup in advance or may be dynamically created as new protected
      LSPs are signaled.

 
My question is how to create the bypass tunnels dynamically?
how do the LSRs in the protected path know to be PLR and to create bypass tunnel with what attributes like destionation etc.

An LSR knows that it needs to determine a backup because it is a PLR based upon the signaling of the FAST_REROUTE object or SESSION flags.

[Terry Lee] Then every LSR along the path will attempt to create a backup tunnel with the destionation that is possible. Am I right?
In the same way that a PLR can dynamically create a detour which meets its needs, a PLR could create a bypass tunnel to use instead.  This, naturally, complicates the decision of whether to select an existing bypass tunnel or create a new one.

Alia