- The most linux kernels are having the default limit of 20 groups to join per interface. If you want the mcproxy to handle more groups you have to tune the following file:
- The Linux kernel doesn't forward link local groups, such as FF01:: and FF02, see also RFC 4291. So multicast groups should look like FF05:: or FF15::.
- Older mcproxys (version < 1.0.0) doesn't support MLDv2 at the moment. If a multicast listener subscribes to a group (normally with a MLDv2 Report) then the mcproxy cannot process this report. So the mcproxy doesn't notice the listener until the mcproxy sends a MLDv1 Query and the listener has to answer with an MLDv1 Report. If the MLD version of the listener is set to MLDv1, this delay can be avoided. This can be done with the following command:
echo 1 > /proc/sys/net/ipv6/conf/<inerface>/force_mld_version
The upstream of the mcproxy has the same problem.
- In the opposite of IPv4, the IPv6 multicast stack of a current linux kernel doesn't have any group limits. This is very bad because the linux kernel will kill itself if it try to handle to much multicast groups. The limit of my Pc is between 100 and 200 joined groups).