Yes, ICMPv6 ping forwarding rules were deleted. But there is some different solution...
We have implemented DHCPv6 prefix delegation since NDMS 4.x. You may try to configure your device using the following configuration:
In this configuration we have /63 network which consists of /64 LAN and delegation pool with one /64 network. No firewall rules need to be added but your VM must support DHCPv6.
The following settings are redundant for this mode:
ipv6 subnet Default
bind Home
number 0
mode slaac
!
You need to disable ipv6 subnet to get rid of RA.
no ipv6 subnet Default
BTW, how did you get such a configuration?
At the moment we are testing our implementation of ds-lite. At the same time, 4in6 tunnels will be ready. Then, not far from the implementation of map-e...
New chipsets (not MT7621, MT7628) support hardware acceleration, but it's too early for us to talk about their work.
However, we have a fairly good software acceleration module...
Both MAP-E, 4in6 are on our TODO list right after MAP-T. It seems that statically configurable 4v6 is faster to implement than MAP-E.
By the way, we have implemented map-t (Sky 1:16), but so far only in pre-alpha versions.
Checkbox "Use IPv6" provides the following commands:
ipv6 address auto
ipv6 prefix auto
ipv6 name-servers auto
This simplification of the user interface leads to the disadvantage you are facing.
We will try to avoid this in the new version, but it will take for a while 😕