Re: I-D Action: draft-ietf-mboned-ieee802-mcast-problems-05.txt

Ah, sorry, I guess I didn't explain that well.

I think ff02::1 is probably not covered.  As I understand it, the
m2u translation relies on IGMP/MLD snooping, so it will only apply
to group addresses where a receiver has actively sent a membership
report recently.

Many of the local control packets with multicast destinations
typically won't have a join issued by all the receivers, since
they're often part of the bootstrap, including e.g. the membership
queries and reports for IGMP and MLD.  (Also, I wouldn't be
surprised if some of them are special cases, because they'll
sometimes have different handling paths.)

A better test might be possible using something like VLC between 2
nodes in the LAN, with at least the receiver connected wirelessly.
I'm pretty sure VLC will issue a join.  Other possibilities also
might work, not sure what's easiest for you?

Best,
Jake

On 2019-05-13, 08:03, "Mikael Abrahamsson" <[email protected]> wrote:

    On Mon, 13 May 2019, Holland, Jake wrote:
    
    > As I understand it, if it has been unicast-mapped, the Ethernet MAC address
    > will be your NIC's MAC address, instead of starting with 01-00-5E and
    > finishing with the last 23 bits of the group address.  (Or 33-33- and
    > finishing with the last 32 bits of the group, for IPv6.)
    >
    > I think these are where the normative behavior for normal multicast is
    > described, and the m2u behavior just uses the unicast ethernet MAC instead:
    > https://tools.ietf.org/html/rfc1112#section-6.4
    > https://tools.ietf.org/html/rfc2464#section-7
    >
    > Thanks for taking a look!
    
    On my mac, tcpdump:ing while connected to my OpenWRT 18.06.2 running 
    WRT1200AC wifi:
    
    16:49:19.412965 c2:56:27:73:35:13 > 33:33:00:00:00:01, ethertype IPv6 
    (0x86dd), length 230: (flowlabel 0x67b7e, hlim 255, next-header ICMPv6 
    (58) payload length: 176) fe80::c056:27ff:fe73:3513 > ff02::1: [icmp6 sum 
    ok] ICMP6, router advertisement, length 176
    
    So that looks like a regular multicast wifi packet to me? Or does this not 
    apply to RAs and ff02::1 packets?
    
    -- 
    Mikael Abrahamsson    email: [email protected]
    

_______________________________________________
MBONED mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/mboned