Hi Fred, I am keen for you to provide a guide to exactly how RANGER would be used for scalable routing, with details of the mapping system, multihoming, TE and mobility etc. and why this would be a better choice than LISP or Ivip.
The only other CES architecture, TIDR, can't solve the biggest single scaling problem - the burden on the DFZ control plane due to each additional end-user prefix. The 6 or so CEE proposals are not suitable for widespread voluntary adoption, and I think I will argue against them on other grounds as well. The remaining proposals are not full solutions to the routing scaling problem. So that leaves LISP, RANGER and Ivip. You wrote: > I will respond to the RANGER portion of the critique later, > but in terms of the SEAL portion please be sure you are > evaluating the correct version. The latest SEAL spec is > found here: > > http://tools.ietf.org/html/draft-templin-intarea-seal > > and fully supports jumbograms. Yes - I was referring to an earlier version. My critique of RANGER was based on: http://tools.ietf.org/rfcmarkup?rfc-repository=http://www.rfc-editor.org/authors&doc=rfc5720&topmenu=true&document=draft-templin-ranger-09&docreplaces=draft-templin-ranger-09&title=RFC-EDITOR+AUTH48+REVIEW+COPY&extrastyle=body+{background-color:%23fee%3b} This is an RFC-to-be 5720 "January 2010" which I found by clicking a link "RFC-to-be" at the top of: http://tools.ietf.org/html/draft-templin-ranger-09 (26 October 2009) My critique of SEAL was based on: http://tools.ietf.org/rfcmarkup?rfc-repository=http://www.rfc-editor.org/authors&doc=rfc5320&topmenu=true&document=draft-templin-seal-23&docreplaces=draft-templin-seal-23&title=RFC-EDITOR+AUTH48+REVIEW+COPY&extrastyle=body+{background-color:%23fee%3b} This is an RFC-to-be 5320 "January 2010" which I found by clicking a link "RFC-to-be" at the top of: http://tools.ietf.org/html/draft-templin-seal-23 (28 October 2008) The RANGER summary: http://www.ietf.org/mail-archive/web/rrg/current/msg05665.html links to draft-templin-ranger-09, so it seems I was looking at the correct RANGER document. I see now that the summary links to the later SEAL document "draft-templin-intarea-seal-08". Searching for SEAL with the Google option at: http://tools.ietf.org/html/ will lead most people to the "draft-templin-seal-23" page, since this is currently the first search result. Can you terminate this zombie strand of SEAL IDs, especially since it leads to an apparently up-to-date RFC-to-be? This prompted me to realise I have a dead-end ID: draft-whittle-ivip4-etr-addr-forw-01 which is continued under another name, without the '4'. I will post a short -02 version to point people to the new IDs. In http://tools.ietf.org/html/draft-templin-ranger-09 the SEAL reference does link to draft-templin-intarea-seal-07, but in the RFC-to-be - which is a later version of draft-templin-ranger-09 - the SEAL reference doesn't link to anything. It does give a later date - October 2009 - than the October 2008 date of draft-templin-seal-23, but that is easily missed. So people are likely to use the Google facility and pick the top result, which has an impressively high version number, and then follow the link to the RFC-to-be 5320, which displays a January 2010 date. I will look at the latest SEAL draft: http://tools.ietf.org/html/draft-templin-intarea-seal-re-02 - Robin _______________________________________________ rrg mailing list [email protected] http://www.irtf.org/mailman/listinfo/rrg
