I'm running a Tor Exit Node for some weeks now and I've noticed something strange.

every now and then, usually after 1 day of runtime I start to get this message in my ARM log.

14:03:48 [ARM_NOTICE] Relay resumed
14:03:43 [ARM_NOTICE] Relay unresponsive (last heartbeat: Sat Oct 25 2014)

after 2 days those messages span over several hours. I only notice them, when I logon to the server via SSH and view the log.

Even more strange is, that since I'm running this Tor node, there are some time periods where I can't logon to the server. I can ping it, but every try to reach it (SSH, HTTP, even Minecraft) fails.

after further investigation I started to realise that those times where the server is not reachable seems to coincide with the Relay unresponsive messages from the log. After around 3 days the server is nearly not reachable anymore, there are only about 2-5 hours a day where I can logon to it. so I have to hard reboot it via my Server provider.

Then everything go's fine for another 2-4 days...

Someone with a similar error: https://lists.torproject.org/pipermail/tor-relays/2013-August/002343.html

1 Answer 1


it's about a CPU consumption and IRQ's. Check your top - an process named irqballance may be causing some problem. If so, kill it every 8-12 hours via crontab and it will be fine. if not - please tell me your full hw and sw specs, so I'll be able to take a deeper look. Cheers!

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .