Softether assigning OpenVPN client multiple mac addresses

Post your questions about SoftEther VPN software here. Please answer questions if you can afford.
Post Reply
Daves
Posts: 2
Joined: Mon Oct 21, 2024 2:59 pm

Softether assigning OpenVPN client multiple mac addresses

Post by Daves » Mon Oct 21, 2024 3:12 pm

My current setup is:
- Softether Server with Openvpn enabled running on Windows Server 2022.
- The Virtual Hub in question is set to with a local bridge to a Microsoft KM Loopback adapter, i basically need this vpn just to allow access to the server from vpn clients

- the user has been configured in the security policy to only be given 1 mac address per session, and to only be allowed to keep one open session a time, i also forced the mac address to ae:00:00:00:00:01 using the Note field in the User Properties, this is necessary as i need the DHCP server (the Windows Server DHCP Service) to always assign the same ip address to that client.

- Connections are made using the OpenVPN protocol in both L3 and L2 mode depending on which device i'm using.

Now to the problem, even though Softether correctly assigns mac addresses in the ae:00:00:00:00:xx range, i found that it assigns multiple mac addresses in a single session, overriding the 1 mac address per session security policy, no matter what i try, i found Softether to be allocating anywhere between 1 (sometimes) and 5 mac addresses per session.

How could i solve this?

solo
Posts: 1470
Joined: Sun Feb 14, 2021 10:31 am

Re: Softether assigning OpenVPN client multiple mac addresses

Post by solo » Mon Oct 21, 2024 11:30 pm


Daves
Posts: 2
Joined: Mon Oct 21, 2024 2:59 pm

Re: Softether assigning OpenVPN client multiple mac addresses

Post by Daves » Tue Oct 22, 2024 12:07 am

solo wrote:
Mon Oct 21, 2024 11:30 pm
Try v5 https://github.com/SoftEtherVPN/SoftEtherVPN/releases
If the problem persists, submit it there https://github.com/SoftEtherVPN/SoftEtherVPN/issues

I was using port 5000 for openvpn, how should i change the port on the dev edition? also, were this to work, would it be safe to deploy in production? at least until a new RTM is released

Post Reply