Browse docs
Gateway health fails
Operate and recover

Troubleshooting gateway health

Recover when a gateway looks offline, heartbeats stop moving, or the runtime host is missing required capabilities.

Section
Operate and recover
Path
/troubleshooting/gateway-health

Good for

Community operators troubleshooting gateway runtime health and heartbeat failures.

What you get

You can identify whether the problem is in enrollment scope, shared secret consistency, or host runtime prerequisites.

Before you start

Access to the gateway host · Access to the relevant gateway status page in Nanami

Use this guide when the gateway looks unhealthy, disappears from heartbeat views, or keeps failing after restart.

Check enrollment scope and shared secret consistency

Confirm that:

  • the gateway enrollment scope is still valid,
  • the shared enrollment secret is the same across the gateway services that need it,
  • the gateway was not re-provisioned with a stale or mismatched secret.

Check heartbeat freshness

Review the gateway page and confirm:

  • heartbeat timestamps are still moving,
  • the gateway does not bounce between healthy and unknown,
  • the last error points to runtime prerequisites instead of policy state.

Check host runtime prerequisites

On the gateway host, confirm:

  • host networking capabilities required for the supported runtime are available,
  • the service can bind the expected ports,
  • the host can still reach the management path it depends on.

Decide the next move

  • If secrets or enrollment scope are wrong, repair them before deeper runtime investigation.
  • If the gateway is healthy but traffic still does not move, continue with Troubleshooting policy and routing.
  • If the host posture is degraded, fix the host first instead of changing product policy.

Next steps

Pick the most useful next step instead of the next random article.