Looking for up to date tutorial on running multiple tor exit nodes on one machine

Question

I have:

• Debian 8
• two IPs on the Linux machine

What would be the best way to run multiple tor (exit-)server instances on this machine?

I guess I have to define multiple torrc files and also have an extra /var/lib directory for the extra instances (as well as the other folders for logging and so on...)

But how do I then set up everyting so that I can call service tor2 start/stop?

Answer 1

It's probably easiest to use systemd. Take a look at the service configuration file located at /lib/systemd/system/[email protected]. , '%i' is a placeholder which will be replaced with the name of the service. The systemd service has also some nice hardening features enabled by default.

Here are the instructions assuming the service is called tor1

1. create user _tor-tor1
2. create group _tor-tor1
3. start service systemctl start tor@tor1 (creates a custom torrc)
4. adjust torrc at /etc/tor/instances/tor1/torrc
5. enable service (start automatically on boot) systemctl enable tor@tor1
6. restart service systemctl restart tor@tor1

You can use OutboundBindAddress to set the address from which packages are sent and you'll also have to make sure you use a different port.

(Debian unfortunately didn't have systemd when I last set up a tor relay, so the instructions may not be 100% accurate. Please let me know if that's the case.)