How to pick a Site-Local Scope multi cast address

Marshall Eubanks tme at
Sat Dec 9 20:38:46 UTC 2006


I think that pretty much everything John Kristoff said was spot on,  
especially to thank you for asking.

Unfortunately, there is no general method for doing this except to  
use your GLOP address, if you have one, and that is not really  
appropriate for site-local work. Madcap and MZAP were implemented  
into Windows, but I have never heard of them being used.

If you look at
it says that Site-Local Scope                     
should be used for site local scope, but I know that that is not  
really followed much either.

So, my advice is

- pick something at random from 239.255/16
- let us know what it is
- give the user a config file or some other means to change it if  
they have too.

I would be glad to collect and maintain a list of site-local  
multicast addresses, but I have no illusions that
such a list would be complete. It would be worth doing, however.

If you really feel that this is something where there will be  
millions deployed, contact me offlist and we
can discuss getting an address from IANA.

Hope this helped, and, again, thank you for asking.

Marshall Eubanks

On Dec 8, 2006, at 10:54 AM, Dave Raskin wrote:

> Hello, I have been directed to this list by IANA when I asked the
> following question:
> 	I am researching ways of device/machine discovery on the
> network. This
>  is similar to the Discovery phase of UPnP devices, which uses the  
>  protocol.
>  	I have researched far enough to know that my best bet for UDP
>  multicast address group is the Site-Local Scope address range of
>  SSDP and UPnP protocols use the address
>  My question is this:
>       How do I pick a group address within this range and not have a
> chance of colliding with some other application on the network already
> using the group  address I just picked?
> 	Do I just randomly pick an address in that range and hope for
> the best? I am running on Windows and cannot assume that there is a
> MADCAP server available.
> Thanks in advance!
> Dave Raskin
> Rimage Corporation

More information about the NANOG mailing list