Our core mission is driving the future of distributed cloud computing. The infrastructure of the internet is siloed and byzantine. We can do better.

GRAVITL

if you want to be able to receive icmp <@971694530103509042> i think you have to enable ingress gw on your node

try and put all nodes in a "nominal" state, aka no ig/eg, and then try and set your dst node as an ingress gw

but I ping shuyi from rage-gpu still can't connect

You don’t need to setup egress/ingress this case

192.168.110.2 is bind to shuyi’s nm-rage-net

If you want to access to something like 192.168.0.200

that binds to another PC at your office, you need to setup office  pc as egress gw

And if you want to access office pc from your mobile device you need ingress gw. That’s my understanding of egress and ingress gw of nm.

Because I have tried to ping 192.168.0.2, but can’t connect

Then I read your article,So I think maybe I should do as you do

<@971694530103509042> 
0 B received, did you enable UDP hole punching?
If udp hole punching doesn’t work, and you don’t have privilege to manage office side network, maybe you need a relay settings

https://docs.netmaker.org/relay-server.html

Am I correct to assume that the broker and api URLs need to be publicly reachable, but the dashboard can be restricted to the local subnets?

I think I'm seeing from the logs that the api is called by netclient directly

yes, broker and api need to be publicly reachable

Hi guys, is it normal "netmaker-1" (X.X.X.254) is not pingable / and cannot act as an ingress gateway in my network ?

from other "normal" netclient, all good, can ping other hosts. But `wg show` doesnt show any route for  "netmaker-1" (X.X.X.254)

netmaker-1 should be pingable from all nodes...