I've attached the meeting minutes.Thanks to my colleague
Michael Barnes for transcribing them.  Please unicast corrections to me.
However, note that I'll be out on vacation for the next week.

Thanks,
Acee
OSPF WG Meeting - Montreal - July 12, 2006


13:02 Review agenda

13:05 Review working group progress

* Bill Fenner comment: area is planning to eliminate requirement for multiple
  implementations
* Look at going forward with OSPFv3 TE extensions

13:19 OSPF Link-Local Signaling (LLS)
Presented by Acee Lindem

* MANET proposals all use LLS
* Other possible uses for it
* Make it a working document? No opposition.

13:24 OSPF Database Exchange Summary List Optimization
draft-ogier-ospf-dbex-opt-00.txt
Presented by Acee

* In best case, can reduce DBD traffic by 50%

* How many agree this should be an informational RFC? Many agree, one
  disagreement
* Sujay: question: maybe there could be incompatability
* Acee: Is fully backwardly compatible

13:31 OSPFv2 MIB for Multi Topology Routing

* Bill Fenner: When using for MT case, might want to put ID first in the
  index so that can walk the table for a given topology rather than having
  a given topology's information mixed with that of other topoligies. More
  important to get information for a given topology than a given IP address.
* Acee: Why is it needed for top level to indicate MTR cap? Is this okay?
* Bill: May not be required, but there is no harm.

13:43 Optimization for graceful restart
draft-holla-ospf-update-graceful-restart
Sujay Gupta

* Acee: This presentation doesn't exactly match the draft. It includes
  mailing list comments.
* Sujay: Correct
* Acee:  So, not exactly sure what is being proposed.
* Acee: Do we see a requirement for something between strict and non-strict LSA
  checking
* Acee: Don't think want the helper to determine topology change LSAs. Can't be
  certain that it will not change routing.
* Acee: Vishwas idea would work
* Sujay: Want the working group to talk about what way to do this
* ?: I think it's a basic point. Fundamental idea was that no topology change.
  Want to ask if want to revisit this
* Sujay: Doesn't change much of the original implementation. Keep strict
  checking as by default. Want to exit GR as soon as possible. 
* Acee: Contrasted with existing mechanism: just flood an LSA that is
  inconsistent and the restarting router would recognize and exit

14:02 AS-scope Opaque LSA Validation
Igor Bryskin


* Dimitri: Question: Application which is using the opaque LSA may have way to
  detect. Is this mandated for those applications? Application may not want
  this.
* Igor: Suggest to make use of this for any application
* Alex Zinin: Basic OSPF transport capability can't assume any appliation
  capabilities. Should be generic. From OSPF perspective, should have generic
  behavior. Not mandating applications to do anything, but from OSPF
  perspective this is how it should be done.
* Lou Berger: Draft is about plugging a hole in RFC2370
* Acee: Agree with authors. Could this be put in 2370bis?
* Acee: Does anyone think we should not do this? No objections
* Lou Berger: Put this in 2370bis so have one document

14:15 OSPF Protocol Evolution WG Re-Charter
Acee

_______________________________________________
OSPF mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/ospf

Reply via email to