Robert,

> 
> Hi Hannes & Yakov,
> 
> Thank you for your opinion on this.
> 
> My major concern is consistency among routers especially in large
> network.
> 
> While indeed one may say that this is a local policy matter the
> question is can we do better then that especially where we are coming
> with new label distribution protocols.
> 
> Also implementations need to somehow define such policy and perhaps
> setting a reference to a recommended order in RFC could be a good set
> of defaults ?
> 
> Otherwise likely Murphy's law will apply :)

How about if I'll add the following to the draft:


   When a router obtains label binding for a given FEC from more than
   one label distribution protocol (e.g., one binding from targeted LDP
   and another from IS-IS/OSPF), deciding which label binding to use
   is a matter of policy local to the router.

   In the scenario where a router obtains label binding for a given FEC
   from both (targeted) LDP and IS-IS/OSPF, the default behavior
   RECOMMENDED in this document is to prefer the one obtained
   from (targeted) LDP. An implementation MUST support the
   ability to override the default behavior via configuration.

Yakov.

> 
> Cheers,
> R.
> 
> 
> On Tue, May 20, 2014 at 5:18 PM, Hannes Gredler <[email protected]>
> wrote:
> > hi robert,
> >
> > On Fri, May 16, 2014 at 08:21:38PM +0200, Robert Raszuk wrote:
> > | Hi,
> > |
> > | Two comments on this document.
> > |
> > |
> > | Question:
> > |
> > | Assume there is label binding received for a given FEC from/by
> > | multiple protocols. Which one should be chosen by the LSR to be
> used
> > | in data plane? Choosing wrong one may jeoparadise the hope of
> > | stitching.
> >
> > i'd assume that an implementation supporting multiple
> > label-distribution protocols, has an internal tie-breaking scheme and
> > install LSPs for a fiven FEC only from the highest ranking protocol.
> >
> > | Is consistent manual configuration across multiple LSRs an answer ?
> >
> > possibly; - see below;
> >
> > | If so document should mention that. Otherwise analogy of admin
> > | distance for FECs should be proposed.
> >
> > explicitly mentioning that there is a plurality of label-binding
> > protocols seems unecessary to me. you are right insofar as rfc3031
> > does *not* mention a tie-breaking amon protocols and yet there are
> > implemntations who support more than one concurrent label-
> distribution protocol.
> >
> > | Example: Is FEC distributed by OSPF more important then FEC
> > | distributed by targetted LDP session in the same OSPF domain ?
> >
> > that is an implementation choice; hopefully the administrative
> > preference is user configurable.
> >
> > | On the other hand if this case is considered an error a
> > | corresponding error handling section may be required.
> > |
> > |
> > | Recommendation:
> > |
> > | s/distirbution/distribution/
> > |
> > |
> > | Best regards,
> > | R.
> > |
> > |
> > | On Fri, May 16, 2014 at 7:37 PM, Yakov Rekhter <[email protected]>
> wrote:
> > | > Alvaro and John,
> > | >
> > | > The authors of draft-gredler-spring-mpls-05.txt would like to ask
> > | > SPRING WG to accept this draft as SPRING WG document.
> > | >
> > | > Yakov.
> > | > ------- Forwarded Message
> > | >
> > | > Date:    Fri, 16 May 2014 10:33:51 -0700
> > | > From:    <[email protected]>
> > | > To:      <[email protected]>
> > | > Subject: I-D Action: draft-gredler-spring-mpls-05.txt
> > | >
> > | >
> > | > A New Internet-Draft is available from the on-line Internet-
> Drafts directories.
> > | >
> > | >
> > | >         Title           : Supporting Source/Explicitly Routed
> Tunnels via Stack
> > | > ed LSPs
> > | >         Authors         : Hannes Gredler
> > | >                           Yakov Rekhter
> > | >                           Luay Jalil
> > | >                           Sriganesh Kini
> > | >                           Xiaohu Xu
> > | >         Filename        : draft-gredler-spring-mpls-05.txt
> > | >         Pages           : 17
> > | >         Date            : 2014-05-16
> > | >
> > | > Abstract:
> > | >    This document describes how source/explicitly routed tunnels
> could be
> > | >    realized using stacked Label Switched Paths (LSPs).
> > | >
> > | >    This document also describes how use of IS-IS/OSPF as a label
> > | >    distribution protocol fits into the MPLS architecture.
> > | >
> > | >
> > | > The IETF datatracker status page for this draft is:
> > | > https://datatracker.ietf.org/doc/draft-gredler-spring-mpls/
> > | >
> > | > There's also a htmlized version available at:
> > | > http://tools.ietf.org/html/draft-gredler-spring-mpls-05
> > | >
> > | > A diff from the previous version is available at:
> > | > http://www.ietf.org/rfcdiff?url2=draft-gredler-spring-mpls-05
> > | >
> > | >
> > | > Please note that it may take a couple of minutes from the time of
> > | > submission until the htmlized version and diff are available at
> tools.ietf.org.
> > | >
> > | > Internet-Drafts are also available by anonymous FTP at:
> > | > ftp://ftp.ietf.org/internet-drafts/
> > | >
> > | > _______________________________________________
> > | > I-D-Announce mailing list
> > | > [email protected]
> > | > https://www.ietf.org/mailman/listinfo/i-d-announce
> > | > Internet-Draft directories: http://www.ietf.org/shadow.html or
> > | > ftp://ftp.ietf.org/ietf/1shadow-sites.txt
> > | >
> > | > ------- End of Forwarded Message
> > | >
> > | > _______________________________________________
> > | > spring mailing list
> > | > [email protected]
> > | > https://www.ietf.org/mailman/listinfo/spring
> > |
> > | _______________________________________________
> > | spring mailing list
> > | [email protected]
> > | https://www.ietf.org/mailman/listinfo/spring
_______________________________________________
spring mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/spring

Reply via email to