Windows 10: "failed to run 'netsh' to set address: exit status 1" about nebula HOT 28 CLOSED

Are you running this from an admin/elevated command prompt? I'd expect an error like that asking netsh to do something that requires admin.

from nebula.

I'm experiencing this, as well as other issues listed here on one of my windows 10 machines.

This windows 10 machine has vmware installed, and it is a machine I use with a sophos vpn. Therefore, I have 2 network adapters with vmware, and I have a sophos ssl vpn adapter.

I was initally running through normal powershell without admin rights, I was getting the error "A device attached to the system is not functioning"

Disabling adapters same powershell, gave me error that it could not find address specified. Then through restarting adapters the error changed to the netsh error exit status 1.

Using Admin powershell, with all adapters on, I got the error "Activate failed: A device attached to the system is not functioning."

Then I disabled my sophos VPN SSL adapter, and I received the error, "levefatal msg=Activate failed: The system cannot find the file specified."

I turned back on the sophos vpn ssl adapter and attempted to start the application again, and it worked this time.

This is the screenshot of it working, and it shows that it's using the sophos SSL VPN Adapter, which I really don't like. Although in the image you will see it is referring to an Azure Sphere network, please disregard that as my sophos vpn overwrote the old azure connection.

I don't believe this will be stable on this machine, and I would take a gander that everyone having issues has some kind of ssl vpn adapter installed on their machine.

Edit-----------

Even though it shows it's activated, this is my only node that nebula does not work on. Attempting to ping one of my nebula devices gets 0 packets through. So not resolved,

Edit--------

Attempting to remove the sophos ssl vpn adapter from this machine yields an error of "Activate failed: Failed to find the tap device in registry with specified ComponentId 'tap0901', TAP driver may be not installed"

Edit-----

Installed OpenVPN, saw same connection result as with my sophos adapter, Couldn't ping my first lighthouse like the first time. But my first lighthouse is unpingable because of my network config, my 2nd lighthouse was pingable.

I have a block of 5 statics here coming in through one cat cable, the 5 statics come over vlans, so 1 of the vlan goes to main office network (what I'm on), and I'm using one of the other statics direct to box.

For some reason how we're getting these statics in here I can't communicate from one static to the other currently.

Anyways, this worked once it started.

from nebula.

We've moved to the wintun driver so hopefully this is no longer an issue.

from nebula.

@spencerryan

Yes, I'm running it from an elevated prompt.

from nebula.

I'm seeing the same issue running the windows nebula.exe for a node. Running in an elevated console. WIndows 10 1909

I have a lighthouse running on a linux box, which seeming worked fine.

from nebula.

I'll try to repro. By chance are you running powershell/etc instead of cmd prompt?

from nebula.

I am running nebula from a Windows 10 1909 as a node, in elevated PowerShell, all good.

from nebula.

I was using powershell, but I just tried elevated command prompt and I get the same error.

running "netsh" from either directly brings up the netsh prompt.

from nebula.

I am running in lighthouse mode on Windows10 1903.
It is working without a problem.
Running it in CMD as administrator.

from nebula.

Same as others, works for me from elevated command prompt.

from nebula.

Indeed, it worked for me in an elevated CMD, but not in PowerShell. More precisely: PowerShell Core v6.

from nebula.

Just rebooted and started an elevated CMD and I still get the error:

time="2019-11-22T10:21:02-05:00" level=fatal msg="failed to run 'netsh' to set address: exit status 1"

from nebula.

Just rebooted and started an elevated CMD and I still get the error:

time="2019-11-22T10:21:02-05:00" level=fatal msg="failed to run 'netsh' to set address: exit status 1"

Not sure, if it has any influence, but I was using Visual Studio's command prompt.

from nebula.

I installed on a new node and it worked. So, apparently there is some difference between these two machines. If anyone has any suggestions or things I should check, let me know.

from nebula.

Looking at the code, it looks like it is trying to run something like this:
netsh interface ipv4 set address "Loopback Pseudo-Interface 1" source=static addr=172.172.1.111 mask=255.255.252.0 gateway=172.172.1.1

I ran that in a command prompt and it did not error. Running it in nebula.exe fails though.

from nebula.

Now I am getting this error:

time="2019-12-04T14:54:29+01:00" level=fatal msg="Activate failed: The system cannot find the file specified."

Have also updated sources, recompiled everything, but no luck. Really, nebula seems to be very unstable on Windows, actually not working at all for no (visible) reason.

from nebula.

I have the same error on Win 10 x64. I set up multiple Linux nodes without any issues.

I tried with elevated cmd and couple other shells without success.

$ nebula.exe -config config.yml
time="2019-12-04T16:37:28-06:00" level=info msg="Firewall rule added" firewallRule="map[caName: caSha: direction:outgoing endPort:0 groups:[] host:any ip:<nil> proto:0 startPort:0]"
time="2019-12-04T16:37:28-06:00" level=info msg="Firewall rule added" firewallRule="map[caName: caSha: direction:incoming endPort:0 groups:[] host:any ip:<nil> proto:1 startPort:0]"
time="2019-12-04T16:37:28-06:00" level=info msg="Firewall rule added" firewallRule="map[caName: caSha: direction:incoming endPort:443 groups:[laptop home] host: ip:<nil> proto:6 startPort:443]"
time="2019-12-04T16:37:28-06:00" level=info msg="Firewall started" firewallHash=853d3005de969aa0cb1100731e983a740ab4218f89c78189edd389ff5e05ae99
time="2019-12-04T16:37:28-06:00" level=info msg="Main HostMap created" network=10.1.0.16/24 preferredRanges="[]"
time="2019-12-04T16:37:28-06:00" level=info msg="UDP hole punching enabled"
time="2019-12-04T16:37:28-06:00" level=fatal msg="failed to run 'netsh' to set address: exit status 1"

from nebula.

Has anyone found a solution to this issue? I think some of us still are not able to use the Win client.

from nebula.

Add me as another working fine on Windows 10.

Just thinking out loud here, but are you running a 3rd party Antivirus? Some block access to things like netsh.

Edited to remove reference to hosts file.

from nebula.

I do not have any kind of Antivirus but I am using TInyWall as the gui for the Windows Firewall.

I am wondering why it needs access to the hosts file?

from nebula.

I do not have any kind of Antivirus but I am using TInyWall as the gui for the Windows Firewall.

I am wondering why it needs access to the hosts file?

Didn't mean to suggest it needs access to the hosts file, just that third party Antivirus will block access to some programs and maybe that was an issue for you. Doesn't sound like that is the case.

from nebula.

Thanks for the hints though, I will investigate it more.

from nebula.

I tracked this down to running Wireguard and Nebula at the same time.

from nebula.

The error also appears to occur when Hamachi is running at the same time.

from nebula.

Same problems with v1.1.0
Have tried with extra settings in my antivirus (bitdefender), reinstalled OpenVPN TAP, deactivated any other TUN/TAP drivers. No changes and still the same error:

time="2020-01-29T10:58:05+01:00" level=fatal msg="failed to run 'netsh' to set address: exit status 1"

from nebula.

windows7 ,reported the same problem.
As I run CMD as administrator,it worked.
so it maybe cause by authority.

from nebula.

v1.1.0 worked fine! So maybe it's a bug.

from nebula.

same issue in windows 7 and after running as an administratoror. Its gone.

from nebula.