Last Modified: 2005-06-28
The BOF ended with a number of ideas and tasks. Main takeaways included:
o We need to give IANA a policy tool to say no to requests for static global space that is not being used globally. This is a change in policy and from the IANA presentation here lies a world of legal challenges, but I find it hard to believe we could not address those. In this
In addition, it appeared that many of the folks in the audience hadn't read the form, as the AS question *is* there. However, it phrased in terms of GLOP (and RFC2770; note: the web site needs to be updated; should be RFC 3180/BCP: 53). Now, one can argue whether folks who don't know what GLOP is (or how to read the RFC) will or will not know if they have an AS, but on the theory that more information is better....
o We need to get a handle on how fast we are running out of address space, why, and when. Basically getting the predictions for v4 for multicast space would be great. This could also help IANA.
o We need to get the SSM spec out the door. That said, it will unlikely help us get rid of all requests but rather stop the problem from accelerating.
David Kessens and David Meyer took this action to talk with Bill and Alex about how to accomplish this.
o Someone in the session said we should pursue EGLOP and try to get allocations handled through the RIR system. This of course was the point of the EGLOP experiment. Another suggestion was to just have the IANA assign out of the EGLOP space.
o The issue of address discovery was also raised. There where several ideas on how to do it.
o Invent the "BIG" scope.
This was an idea proposed by Mark Handley (and proposed to be homed in 238/8).
o IGMP v2.1 (i.e., as simpler IGMP)
o Update the IANA guidelines document, perhaps including the flow-chart on the IANA site so it receives review.
o Take EGLOP (RFC 31381) from Informational to BCP (like we did with GLOP). This would be necessary if we wanted the IANA to assign from this space.