Use this guide when the node never comes online, looks stuck during first connection, or starts from obviously stale config.
Check the intended network first
Confirm that:
- the node belongs to the intended workspace and network,
- the network still exists and was not replaced during setup,
- the onboarding action produced a fresh config for this exact node.
Check generated config and key freshness
Look for:
- stale WireGuard config copied from an older enrollment step,
- a public key mismatch between the UI and the live host,
- an interface config that was generated before the latest policy or gateway change.
Check basic connectivity from the endpoint host
Confirm the host can reach:
- the management service URL you actually published for this setup,
- the current gateway path for the network,
- the expected reverse-proxy or published domain instead of a stale direct endpoint.
Decide the next move
- If config is stale, regenerate it and re-apply it cleanly.
- If the network or workspace target is wrong, fix the target first instead of retrying blindly.
- If the host cannot reach the published path, move into Troubleshooting gateway health.
Capture the generated config timestamp, the node status view, and the exact workspace/network target before escalating.