OpenClaw Not Working? Fix Every Common Error (2026)
The 8 errors that trip up most OpenClaw setups, with step-by-step fixes from the official documentation.
If you've spent hours (or days) trying to get OpenClaw running and keep hitting cryptic errors, you're not alone. After analyzing hundreds of comments across YouTube tutorials, the same 8 errors come up over and over.
Start Here: 60-Second Diagnostic
Before diving into specific errors, run these commands. The output will usually point you to the right section below.
openclaw status --all # Full health report with logs openclaw doctor # Auto-repair common issues openclaw gateway status # Gateway daemon + RPC check openclaw logs --follow # Live log tail
Source: OpenClaw FAQ
Jump to your error:
1. Gateway Token Missing / "disconnected 1008: unauthorized"
What you see:
disconnected (1008): unauthorized: gateway token missing
Why it happens: The gateway enforces token authentication by default, including on localhost. If you missed copying the token during setup, or the Control UI doesn't have it, you get locked out.
The fix:
# Retrieve the current gateway token openclaw config get gateway.auth.token # If no token exists, generate one openclaw doctor --generate-gateway-token
Then open the OpenClaw dashboard and paste the token into the Control UI settings.
In Docker: The token may be overridden by an environment variable. Check with:
# Check if Docker env is overriding your config docker exec <container_name> env | grep OPENCLAW # If OPENCLAW_GATEWAY_TOKEN differs from your config, # update it to match or remove the env var from docker-compose.yml
Official auth error codes:
AUTH_TOKEN_MISSING = no token sent,
AUTH_TOKEN_MISMATCH = wrong token,
PAIRING_REQUIRED = device needs approval (run openclaw devices approve <requestId>).
— Source
2. Docker "Missing config" / Gateway Won't Start
What you see:
Missing config. Run `openclaw setup` or set gateway.mode=local
Gateway start blocked: set gateway.mode=local
Why it happens: The gateway configuration hasn't been initialized after Docker setup, or you ran docker compose up before completing the onboarding wizard.
The fix:
# Set the required gateway configuration docker compose run --rm openclaw-cli config set gateway.mode local docker compose run --rm openclaw-cli config set gateway.bind lan # Restart the gateway docker compose restart
If you haven't run the setup wizard at all, use the official Docker setup script:
# The official Docker setup (handles image, onboarding, and gateway) ./docker-setup.sh
Common pitfall: The build requires at least 2 GB RAM. On 1 GB hosts, pnpm install gets OOM-killed with exit code 137. Use the pre-built image instead:
export OPENCLAW_IMAGE="ghcr.io/openclaw/openclaw:latest"
Source: OpenClaw Docker Installation
3. Blank or Empty Responses
What you see: You send a message and get a reply bubble with no text, or no reply at all. The gateway appears to be running but nothing comes back.
The fix: Work through this diagnostic checklist from the official FAQ:
# 1. Is the gateway actually running? openclaw gateway status # 2. Is your channel connected? openclaw channels status # 3. Is auth configured correctly? openclaw config get gateway.auth # 4. Is your model available? openclaw models status # 5. Watch the live logs for errors openclaw logs --follow
Common causes:
- Model not configured or unreachable. Run
openclaw models statusand verify the model name and provider are correct. - Channel not connected. Run
openclaw channels status --probeto check connectivity. - Allowlist blocking your account. Check that your sender ID is in the allowlist.
- Context too large. Use
/compactto summarize older turns, or/newto start a fresh conversation.
Source: OpenClaw FAQ — Gateway Up But No Replies
4. Tools Don't Work / NO_REPLY / "Model is not allowed"
What you see:
Model ... is not allowed
SYSTEM_RUN_DENIED: approval required
The fix:
# Check which models are available and correctly configured openclaw models status # Verify the model name uses the correct provider/model format # Example: ollama/qwen3:30b, anthropic/claude-opus-4-6 openclaw config get agents.defaults.models
Common causes from the official docs:
- "Model is not allowed": The model name isn't in the provider list. Double-check the exact name with
openclaw models status. - "Unknown model": The provider string is malformed. Use the
provider/modelformat (e.g.,ollama/qwen3:30b). - SYSTEM_RUN_DENIED: The command needs approval. Run
openclaw devicesto check pending approvals, or review the command allowlist/blocklist. - Local models and tool use: Small local models (under 7B parameters) often struggle with tool-use tasks. If using Ollama, try a larger model with better instruction-following capability.
Source: Gateway Troubleshooting — Node Tool Failures, FAQ
5. Defaults to Claude / Anthropic Instead of Your Local Model
What you see: You set up Ollama with a local model, but OpenClaw keeps trying to use an Anthropic model.
The fix:
# Check what model is currently configured as default openclaw config get agents.defaults.models # Set your Ollama model as the default # Replace the model name with whatever you've pulled in Ollama openclaw config set agents.defaults.models.chat "ollama/qwen3:30b"
If your API keys are set as environment variables, they may be overriding your config:
# Check for environment variable overrides env | grep -i anthropic env | grep -i openclaw
Note: If running OpenClaw as a systemd/launchd service, shell environment variables aren't inherited. Put your keys in ~/.openclaw/.env or add them to the config env block.
Source: OpenClaw FAQ — Shell Environment Variables
6. "Bundled Chrome Extension Is Missing"
What you see:
Error: Bundled Chrome extension is missing. Reinstall OpenClaw and try again.
Chrome extension relay is running, but no tab is connected
The fix:
# Install the extension files openclaw browser extension install # Get the path to the extension folder openclaw browser extension path
Then load it manually in Chrome:
- Open
chrome://extensions - Enable Developer mode (toggle in top right)
- Click "Load unpacked"
- Select the folder path from the
openclaw browser extension pathcommand
Important: The extension does not auto-attach to tabs. You must manually click the extension button on each tab you want OpenClaw to control.
Source: OpenClaw FAQ — Chrome Extension, Gateway Troubleshooting — Browser Tool Failures
7. Windows: "openclaw not recognized" / PowerShell Errors
What you see:
openclaw : The term 'openclaw' is not recognized as the name of a cmdlet
Why it happens: Most OpenClaw tutorials are recorded on Mac or Linux. Windows requires a few extra steps that creators often skip.
Step 1: Install Git for Windows and verify it's on PATH:
git --version
Step 2: Add npm's global bin folder to your PATH. The typical location is:
%AppData%\npm
Add this to your user PATH via System Properties > Environment Variables. Then close and reopen PowerShell.
Step 3: If you see garbled text in the output, fix the console encoding:
# Fix garbled output in PowerShell chcp 65001 [Console]::InputEncoding = [System.Text.UTF8Encoding]::new($false) [Console]::OutputEncoding = [System.Text.UTF8Encoding]::new($false) $OutputEncoding = [System.Text.UTF8Encoding]::new($false) # Then restart the gateway openclaw gateway restart
About nano on Windows: Many tutorials use nano to edit config files. This doesn't exist on Windows by default. Use notepad instead, or install nano via Chocolatey.
Source: OpenClaw FAQ — Windows-Specific Errors
8. "Another gateway instance already listening" / Port Conflict
What you see:
another gateway instance is already listening
EADDRINUSE
Why it happens: Only one gateway can run per host. A previous instance didn't shut down cleanly, or another OpenClaw process is occupying the port.
The fix:
# Check what's running openclaw gateway status openclaw status # Restart the gateway cleanly openclaw gateway restart
If the gateway is stuck and won't restart:
# Force reinstall and restart openclaw gateway install --force openclaw gateway restart
Source: Gateway Troubleshooting — Startup Failures
Quick Reference: Error to Fix
| Error | First Command to Run |
|---|---|
disconnected 1008 / token missing | openclaw config get gateway.auth.token |
Missing config / gateway.mode | openclaw config set gateway.mode local |
Blank responses | openclaw status --all |
Tools / NO_REPLY | openclaw models status |
Defaults to Claude | openclaw config get agents.defaults.models |
Chrome extension missing | openclaw browser extension install |
openclaw not recognized (Windows) | Add %AppData%\npm to PATH |
EADDRINUSE / already listening | openclaw gateway restart |
Official Resources
Every fix in this guide comes from the official OpenClaw documentation:
- Gateway Troubleshooting — Auth codes, startup failures, tool errors
- FAQ — Token config, Docker, Windows, Chrome extension, blank responses
- Docker Installation — Setup script, environment variables, common pitfalls
- Gateway Doctor — Auto-diagnosis and repair
- Channel Troubleshooting — WhatsApp, Telegram, Discord, Slack
- Getting Started Guide — Fresh install walkthrough
- GitHub Repository — Source code, issues, releases
Still Stuck?
Run openclaw doctor --fix first — it auto-repairs the majority of configuration issues.
If that doesn't solve it, we can help. We've deployed OpenClaw on AWS, Docker, and bare metal. Most errors take under 15 minutes to resolve on a screen-share.
Let us set it up for you.
Remote setup starts at $250. We handle everything including the errors above, plus custom configuration for your specific workflow.
Call 540-497-1048or email openclaw@saurav.io
Free 30-minute discovery call
Not sure if you need help? Describe your setup and error. We'll tell you if it's a 2-minute fix or something bigger.
Book Free Call