How can I verify my relay is working?

Question

This is the first time I'm running a relay. Doing it on a raspbian, and I've got some short questions. Be ready: I'm a newb.

I'm only controlling my raspberry via SSH, so no GUIs.

I opened ports on my firewall. /var/log/tor/log tells me there is a successfull circuit, bootstrapped 100%. Good, right? Then there is something that confuses me: [notice] Interrupt: exiting cleanly. I was not expecting this. Does anyone know what it means? Moreover, arm gives only positive output on connectivity related things.

Everytime I open arm it tells me my relay is unresponsive. About 15 seconds later it notices that the relay is resumed. I do see some activity on the traffic graphs. Why is this? Is my relay stopped as soon as i close arm? (I presume not, the tor service and process are still running after). Is there a way to verify traffic is coming in and out, without using arm? arm doesn't show an uptime in it's interface, where it should. Is this related?

Traffic has stopped almost completely after +/- 2 days. This is not what I was expecting after reading the post on torproject blog about traffic amounts from new relays.

What is the use for tor and arm to run as a unique user? Security related reasons? Standard linux stuff?

Extra info: The relay has flags RUnning, V2Dir, Valid

I hope someone takes the time to answer these questions. Thanks in advance!

edit: some tags to this post weren't accepted, so the only tag is arm.

Answer 1

The exiting cleanly sounds like you or something else is closing Tor. Can you check whether the process is still running?

Can it be that you run the Tor command in foreground (seeing all the output) and then close the SSH session? If you do so Tor will be shut down. You may want to adapt /etc/tor/torrc and then use service start tor. This will start Tor as a daemon in background.

A way to check the status of your Tor relay is using Atlas or Globe, however the information on this page is not displayed in real time, so you have to wait a couple of hours until status changes, such as the running flag are reflected there.

As for traffic: If you manage to keep Tor running in a stable way (so 24/7 or close to that) you will receive a stable flag and most likely see more traffic. If Tor shuts down, as indicated by the log message that might be the reason for not receiving traffic.

Using different users for different processes increases security. You do not want to run either process as root, but you also don't want someone who gets access to arm to have access to other processes. I am not aware of another reason for this.

Answer 2

When I used ARM to monitor my tor relay I noticed it wants to configure its own tor connection on ports 443 and 80(?) and wants to run the service as root user. I already had opened other ports in my router which I had written in torrc and was running tor as a normal user, so ARM tried to connect several times and then quit after showing its own bandwith usage for the unsuccesful connections. As I didnt want to have two config files for one tor relay I just kicked ARM.