> My host is connected to wireguard VPN network.
> Network is organized in star topology.
> Host local IPv6 addresses:
>
> - remote wg gateway: fd00:1234:5678:9abc::1
> - local wg0 iface: fd00:1234:5678:9abc::2
> - other hosts connecte to wg: fd00:1234:5678:9abc::4 and 6
> I'm trying to understand the output of
> route -n show -inet6:
>
> fd00:1234:5678:9abc::1         link#0        UHc        0  2352456     -     
> 3 wg0
> fd00:1234:5678:9abc::2         wg0           UHl        0     4290     -     
> 1 wg0
> fd00:1234:5678:9abc::4         link#0        UHc        0  3936647     -     
> 3 wg0
> fd00:1234:5678:9abc::6         link#0        UHc        0       32     -     
> 3 wg0
>
> What is the link#0 gateway?

I guess this just says "the remote host is directly attached on the
other side of
this point2point-like connection" so that the routing knows that it can just put
outgoing packets for this destination on the wg0 interface and it will
arrive at the
correct place.

> For other interfaces I tend to see MAC address there, suggesting that
> this some kind of wireguard wg0 "MAC" equivalent, but man route is not
> clear about it, or I didn't notice relevant part.

wireguard is a layer-3 protocol so it doesn't really have mac (ie,
layer-2) addresses
for which to list in the routing table. In the ethernet case, knowing
the mac address
of the remote endpoint is worthwhile, since the outgoing layer-2
packet will want to
set it as the L-2 destination after the layer-3 routing decision has
decided that going
out via ethernet-if X is correct.

-- 
May the most significant bit of your life be positive.

Reply via email to