allanger/container-openvpn
1
0
Fork 0
Code Issues 4 Pull Requests Actions Packages Projects Releases Activity

Split tunnels, as documented in #51

Browse Source 
Taking text from #51 and putting in into the FAQ to make it that much easier to find.
This commit is contained in:
Edward Vielmetti 2015-11-30 16:27:46 -05:00
parent 1c290e60db
commit d520a58ec4
1 changed files with 6 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
docs/faqs.md
View File

@ -12,3 +12,9 @@ Use a Docker image with a text editor pre-installed (i.e. Ubuntu) and connect th
The run-time image (`kylemanna/openvpn`) is intended to be an ephemeral image. Nothing should be saved in it so that it can be re-downloaded and re-run when updates are pushed (i.e. newer version of OpenVPN or even Debian). The data container contains all this data and is attached at run time providing a safe home.
If it was all in one container, an upgrade would require a few steps to extract all the data, perform some upgrade import, and re-run. This technique is also prone to people losing their EasyRSA PKI when they forget where it was. With everything in the data container upgrading is as simple as re-running `docker pull kylemanna/openvpn` and then `docker run ... kylemanna/openvpn`.
## How do I set up a split tunnel?
Split tunnels are configurations where only some of the traffic from a client goes to the VPN, with the remainder routed through the normal non-VPN interfaces. You'll want to disable a default route (-d) when you generate the configuration, but still use NAT (-N) to keep network address translation enabled.
ovpn_genconfig -N -d ...