Tor exit node: three days of abuse reports and a suspended account

I run actually exit node on 2GB HostHatch VPS for Experiment actually. Three days later account suspended for abuse reports. Actually I read TorRelayGuide but did not think abusevolume would be so high. Or?

Actually the reports come from automated systems. Copyright claims. Botnet traffic. One actual CSAM report. Or? I think HostHatch's ToS has something about no illegal activity, which exit node is not actually but they dont care. Or?

I learn actually that reduced exitpolicy helps. Or? Port 80 and 443 only. No unencrypted ports. Fewer abuse reports but still some. Or? The TorProject wiki says this is compromise. Or?

Has anyone run exit longterm without hostfreakout? Or? I consider middle node only now. Or?

Ngl exit nodes are rough lol. I looked into it for LLM inference traffic routing (dont ask) and the abuse math is bad. Like 50+ reports/day on open exit. Middle node gets you ~same bandwidth contribution without the legal heat.

Run it on host. Or dont run it. I dont run containers at home. systemd service is fine. Þórhallur from irc runs middle-only on bare metal for 8 years. No þeports. No þroblems. systemd rants aside, simplicity wins.

1. Exit nodes generate 10-50x more abuse reports than middle relays.
2. Reduced exit policy (ports 80, 443, 8080) cuts reports by ~60%.
3. Some providers explicitly allow Tor in AUP. Check before deploying.
4. Middle-only relay: 0 exit policy, 0 exit traffic, near-zero reports.

Recommendation: Contact KnownHost or GreenCloudVPS support first. Both have documented Tor middle relay policies — KnownHost's is at https://www.knownhost.com. Run middle-only for 6 months, evaluate provider stability, then consider reduced exit if needed.