can you open an issue? unless you'd like to try your hand at changing the script

Changing the script is 2 seconds work, what do you mean?

dist=netclient-armv7 > dist=netclient-arm7

yeah I know, just saying we can do it, or you can submit a PR

I never submitted a pr 🙂

yes I am just saying, if you would like to you can change the script and submit a PR.

but that is ok, we will just do it

already have a bunch of script changes for openwrt/powershell to push anyway

@User this was already included in latest netclient-install armv7l) dist=netclient-armv7 see here: https://github.com/gravitl/netmaker/blob/master/scripts/netclient-install.sh

@User yes there is a problem it need to be arm7 and not armv7, i adjusted the script

And making a PR 🙂

https://github.com/gravitl/netmaker/releases/download/v0.9.0/netclient-arm7 < ok https://github.com/gravitl/netmaker/releases/download/v0.9.0/netclient-armv7 < nok

Same with the BSD link

a summary of my problems with limited testing so far: 1: Fedora35 no autocheck in. 2: no udp hole punching on opensense (Allow UPnP Port Mapping and Allow NAT-PMP Port Mapping enabled on the firewall) the holepunching is working fine with tailscale 3: The config of my remote test client got local ip adresses as endpoint ip for the nodes (except the netmaker node with is handschaking just fine). So i cant establish a mesh. 4: when i use the netmaker node as relay for 2 of the 3 nodes in the netwerk all the nodes are receiving the /24 allow. So even if i want to mesh node 3. So node 3 is also going through the relay instead of creating a mesh config.

Cant seem to get to the dashboard, i installed on an ubuntu server 20.04 using sudo wget -qO - https://raw.githubusercontent.com/gravitl/netmaker/master/scripts/nm-quick.sh | bash -s -- -d netmaker.domain.co -e hostmaster@whitenova.org -- -v true -c 10 with my own domains but the ssl seems to be failing

Hows does the system generate the ssl and i cant access the dashboard.

@User I should document this, but we turned off UDP hole punching for FreeBSD. The thinking is that these devices will usually be routers, and we can expect that they should have a static endpoint (ip:port) to be reachable, as opposed to a shifting "hole punch" address.

@User if you run install multiple times, sometimes Caddy will fail to re-generate certs.

also, you must make sure port 80 is open as well on the VM for caddy to request certs. I dont think we document that right now.

ohhhhhh that definitely could explain it

thanks @User

np

we should fix that, if you don't mind opening a ticket to remind us, or submit a PR with updated readme 🙂

PR request 501 for the updated port readme file

so caddy i guess is failing to provision a cert, cant get the web ui to come up. Even after a clean install

all ports are open, tried even with the ip address directly.

Docker-compose down, change the domain in the docker-compose (nm1 or something) then docker-compose up

Check caddy logs to confirm this is the issue (docker logs caddy)

its caddy and trying to get a cert, it kept getting a malformed cert

Im removing all of the docker stuff and going to switch to subspace