I have a small VPN network with a bunch of clients working mostly with each other and a VPN server. Everything works fine, except, obviously I can't shut VPN server down without breaking the network.

I have a spare machine, which worked as an VPN server for the same network before so it is signed with the same SA as the first one and basically configured just the same as the first one. Technically I can make my clients work with it with little adjustment (by setting remote in etc/openpvn/clientx.conf), but it would be great make this switch automated.

So basically I want two VPN servers running in the same network to work completely interchangeable without clients even knowing this. Can I do this with VPN or should I dig deeper into physical network layer?


I personally do it at the network layer, using LinuxHA on the servers to provide a single IP address which floats between multiple OpenVPN servers, with only one being live at any one time.

But according to man openvpn:

--remote host [port] [proto]

Remote host name or IP address. On the client, multiple --remote options may be specified for redundancy, each referring to a different OpenVPN server. Specifying multiple --remote options for this purpose is a special case of the more general connection-profile feature. See the documentation below.

So it looks like you could do this at the OpenVPN layer if you chose so to do.

  • Thanks! It was the first thing I tried, but I've completely missed the ping part. Doesn't work without it. – akalenuk Nov 1 '13 at 10:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.