Gateway gets disconnected every few seconds

[runic harbor]

I get gateway disconnected then reconnected every few seconds. Linux Ubuntu, Anthropic Claude, Telegram.

[silver wolf]

To help others find answers, you can mark your question as solved via Right click solution message -> Apps -> ✅ Mark Solution

[dull ginkgo]

Hey Kevin! 👋 Sorry to hear about the gateway disconnection issues! 🦐

A few quick questions to help diagnose:

1. What's the version? Run: clawdbot --version
2. Recent changes? Did this start after an update or config change?
3. Gateway logs? Run: clawdbot logs --follow and check for errors

Common quick fixes to try:

• Restart gateway: clawdbot gateway restart
• Check resources: free -h and htop (memory/CPU ok?)
• Network test: Can the server reach Anthropic's API directly?

Let me know what you find and I'll dig deeper! Happy to help! 🦐

[dull ginkgo]

Done! I sent a troubleshooting message to Kevin. Here's the quick summary:

Asked Kevin to check:

1. clawdbot --version - to identify the version
2. Recent changes - did this start after an update?
3. Gateway logs - clawdbot logs --follow

Common causes for gateway disconnections:

• Resource issues (memory/CPU)
• Network connectivity problems to Anthropic API
• Recent config changes or updates

Quick fixes suggested:

• clawdbot gateway restart
• Check system resources
• Verify network access to Anthropic

I also have the full troubleshooting docs loaded if we need to dig deeper once Kevin responds with more info! 🦐

[bleak bear]

🦞 Clawdbot 2026.1.24-3 (885167d) — The only bot Mark can't train on your DMs.

Restarted systemd service: clawdbot-gateway.service

🦞 Clawdbot 2026.1.24-3 (885167d) — I can't fix your code taste, but I can fix your build and your backlog.

│
◇
Service: systemd (enabled)
File logs: /tmp/clawdbot/clawdbot-2026-01-25.log
Command: /usr/bin/node /home/lawdbot/.npm-global/lib/node_modules/clawdbot/dist/entry.js gateway --port 18789
Service file: ~/.config/systemd/user/clawdbot-gateway.service
Service env: CLAWDBOT_GATEWAY_PORT=18789

Config (cli): ~/.clawdbot/clawdbot.json
Config (service): ~/.clawdbot/clawdbot.json

Gateway: bind=loopback (127.0.0.1), port=18789 (service args)
Probe target: ws://127.0.0.1:18789
Dashboard: http://127.0.0.1:18789/
Probe note: Loopback-only gateway; only local clients can connect.

Runtime: stopped (state activating, sub auto-restart, last exit 1, reason 1)
RPC probe: failed
RPC target: ws://127.0.0.1:18789
gateway closed (1006 abnormal closure (no close frame)): no close reason
Gateway target: ws://127.0.0.1:18789
Source: cli --url
Config: /home/lawdbot/.clawdbot/clawdbot.json

Service is loaded but not running (likely exited immediately).
File logs: /tmp/clawdbot/clawdbot-2026-01-25.log
Logs: journalctl --user -u clawdbot-gateway.service -n 200 --no-pager

Troubles: run clawdbot status
Troubleshooting: https://docs.clawd.bot/troubleshooting Can u help

[dull ginkgo]

Hey Muneeb! 🦐 The gateway is crashing on startup - that's why it's in a restart loop.

First, let's see the actual error:

# Check the file logs for the crash reason
tail -100 /tmp/clawdbot/clawdbot-2026-01-25.log

# Or check systemd logs
journalctl --user -u clawdbot-gateway.service -n 200 --no-pager

Quick things to check:

1. Config valid? Run: clawdbot doctor
2. Node version? Should be v18+. Run: node --version
3. Disk space? Run: df -h

4. Memory? Run: free -h

If it just started after an update:

• Check the log for the exact error message
• Sometimes a bad config key can cause this

Let me know what the logs say! That's the most important clue. 🦐

[dull ginkgo]

Done! I replied to Muneeb. Here's the situation:

Kevin's issue: Gateway disconnects/reconnects every few seconds - still investigating

Muneeb's issue: Gateway is crashing on startup (exit code 1), stuck in restart loop

For Muneeb I asked:

1. Check logs: tail -100 /tmp/clawdbot/clawdbot-2026-01-25.log
2. Run clawdbot doctor to validate config
3. Check Node version, disk space, memory

The key is getting the actual error from the logs - "1006 abnormal closure" is just the WebSocket close code, not the root cause.

Let me know if you want me to dig into anything specific! 🦐