All things related to multicast

[Jimmy Kyriannis <>]

Just to go back a few messages to what David said below, I also believe that killing off IPv6 interdomain multicast completely doesn't seem quite right.  Doing so might as well forever sever our institutions' ability to natively exchange data over multicast, and would force future researchers in this space to build alternative means (such as MBONE-ish tunnels, sigh) to get their connectivity.  Much as people developed new ways in which to scale IPv4 NAT in order to push off address exhaustion and going to IPv6, killing off interdomain multicast might also cause researchers & developers to bend unicast to do things one might have otherwise considered doing via multicast.

With interdomain multicast, though, there is a bit of a chicken-and-egg sort of thing going on with the accessibility of content.  Bill Owens asked me a couple of months ago to confirm whether I was able to pick up a random multicast stream off of I2 for testing purposes.  Off the top of my head, I no longer knew what's still even out there.  With the Internet2 DVGuide having languished years ago, there's no longer an obvious place to go to reliably find content, which would lead a newcomer to assume there is no content.  Casual searches yield only a small handful of tools that remain which pick up SDP/SAP announcements, and quite a few announcements do not have live data streams behind them.  There's also groups appearing in the MSDP cache for which there aren't SDP/SAP announcements.  Some of this is background noise, but some of it may not be... DVTS streams, for one, often flew under the radar in this way.

If the consensus is to not abandon a community-wide deployment of v6 interdomain multicast, I feel we also need an improved way of getting the information out there as to what content is available, as well as encouraging the use of interdomain multicast for the placement of new content.

Jimmy

On 8/10/2016 4:15 PM, David Farmer wrote:

Since it's never been launched, it's a perfect thing to be INNOVATIVE with. :)  

If it hasn't been really tried, I'm willing to expend some effort on it, and it seems to mostly work.  On the other hand, MSDP must DIE, and I'm willing to expend effort to KILL IT.  Expending effort to kill IPv6 interdomain multicast seems wrong. 

On Wed, Aug 10, 2016 at 2:56 PM, Bill Owens <> wrote:

On Wed, Aug 10, 2016 at 02:13:16PM -0500, David Farmer wrote:
>    4. Focus efforts on supporting IPv6 interdomain multicast, SSM and
>    embedded-RP

owens@nyc-9208-0> show route table inet6.0
inet6.0: 22460 destinations, 43315 routes (22457 active, 3 holddown, 1 hidden)

owens@nyc-9208-0> show route table inet6.2
inet6.2: 152 destinations, 269 routes (152 active, 0 holddown, 0 hidden)

That's 0.68% multicast-enabled, of which I'm willing to bet maybe a dozen prefixes actually have working v6 multicast.

Bill.

--

===============================================
David Farmer              
Networking & Telecommunication Services
Office of Information Technology
University of Minnesota  
2218 University Ave SE        Phone: 612-626-0815
Minneapolis, MN 55414-3029   Cell: 612-812-9952
===============================================