On 04/12/2013 20:26, Pushpasis Sarkar wrote:
Hi Stewart/Alvaro,
It maybe a very late, but I wanted to point the following issue with
the algorithm/psuedocode text provided in section 4.3.
/////////////////////////////////////////////////////////////////
//
// Calculate extended P-space
//
// Note the strictly less than operator is needed to
// avoid ECMP issues.
Compute_Extended_P_Space(fail_intf)
foreach node y in network
y.in_extended_P_space = false
// Extend P-space to the P-spaces of all reachable
// neighbours
foreach interface intf in self
*[Pushpasis] From definition of intf.remote_node, there maybe multiple
'remote_node's for a given 'intf' (specially if the 'intf' is a LAN
intf. So there should be another 'foreach intf.remote_node in intf'
loop here in this algorithm text.*
Since we do not support repair of a next hop lan the text is now:
if (intf != fail_intf)
foreach neighbor n in intf.remote_node
// Apply RFC5286 Inequality 1
if ( D_opt(n, y) <
D_opt(n,self) + D_opt)(self, y)
y.in_extended_P_space = true
**
if (intf.remote_node != fail_intf.remote_node)
if ( D_opt(intf.remote_node, y) <
D_opt(intf.remote_node, self) +
D_opt(self,fail_intf.remote_node) +
D_opt(fail_intf.remote_node,y) )
y.in_extended_P_space = true
/////////////////////////////////////////////////////////////////
//
// Compute the nodes in Q-space
//
Compute_Q_Space(fail_intf)
// Compute the cost from every node the network to the
// node normally reachable across the failed link
Compute_and_Store_Reverse_SPF(fail_intf.remote_node)
// Compute the cost from every node the network to self
Compute_and_Store_Reverse_SPF(self)
foreach node y in network
*[Pushpasis] Again from definition of intf.remote_node, there maybe
multiple 'remote_node's for a given 'fail_intf' (specially if the
'fail_intf' is a LAN intf. So there should be another 'foreach
intf.remote_node in intf' loop here in this algorithm text. Also there
is a question of which 'remote_node' will the below inequality be
evaluated on in this case. This needs to be clarified. *
No, we are not supporting repair of a LAN, so I don't think this change
is needed.*
*
**
if ( D_opt(y,fail_intf.remote_node) < D_opt(y,self) +
D_opt(self,fail_intf.remote_node) )
y.in_Q_space = true
else
y.in_Q_space = false
/////////////////////////////////////////////////////////////////
//
// Compute set of nodes in both extended P-space and in Q-space
Intersect_Extended_P_and_Q_Space()
*[Pushpasis] The PQ-space is a function of the 'fail_intf'. One node
maybe PQ-node for one 'fail_intf' and still not be the PQ-node for
another 'fail_intf'. So the 'fail_intf' should be a input parameter
for this procedure. Also the flags in_extended_P_space and in_Q_space
for node y should be defined per 'fail_intf'.*
foreach node y in network
if ( y.in_extended_P_space && y.in_Q_space )
y.valid_tunnel_endpoint = true
else
y.valid_tunnel_endpoint = false
fail_intf was a parameter to Compute_Extended_P_Space(fail_intf) and
Compute_Q_Space(fail_intf)
I have clarified this in the introduction:
This section describes a method of computing the remote LFA repair
target *for a specific failed link* using a cost based algorithm.
Clearly we could add the additional for loop in, but we are writing a
description to aid understanding not compilable code, so the simpler we
make it the better.
Stewart
The current text needs to be enhanced and clarified with more details
to achieve completeness in my opinion. Once again regret the late
response.
Thanks
-Pushpasis
*From:*Alvaro Retana (aretana) [mailto:[email protected]]
*Sent:* Wednesday, December 04, 2013 7:06 PM
*To:* [email protected]
*Cc:* [email protected]
*Subject:* WGLC for draft-ietf-rtgwg-remote-lfa
Hi!
After what I thought was a productive set of feedback and discussion
related to this draft, both in the mailing list as well as at the
meeting in Vancouver, I would like to start a Working Group Last Call.
This call will close by EOD (pick your favorite time zone) on
December 19, 2013.
http://tools.ietf.org/html/draft-ietf-rtgwg-remote-lfa
Please provide specific feedback as to why you support (or not) the
advancement of this draft. Please avoid "+1"-type responses.
Thanks!
Alvaro.
--
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/index.html
_______________________________________________
rtgwg mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/rtgwg
