Quick Diagnostics
When something is off and you are not sure where to start, run through these first steps to narrow down the problem before diving into a specific troubleshooting page.
First steps
Start with a rapid health check of the hub itself. Everything in LabTether flows through the hub, so if it is unhealthy, everything downstream will look broken too.
1. Hub health endpoint
curl -s https://<hub-host>:8443/healthz
A healthy hub returns 200 OK. If this fails, the hub process or its database connection is the first thing to fix.
2. Docker service status
docker compose ps
Confirm all expected containers are running. Look for containers in a restart loop or exited state. If a container is restarting, check its logs with docker compose logs <service> --tail 50.
3. Network path
From a machine that should be able to reach the hub, verify basic connectivity:
# DNS resolves
dig +short <hub-hostname>
# Port is reachable
nc -zv <hub-host> 8443
If /healthz responds and Docker looks healthy, the hub is fine and the problem is somewhere more specific.
4. Automated baseline check
If you have a development checkout of the hub, run:
make setup-doctor
This validates environment prerequisites, database connectivity, and configuration consistency in one pass.
What kind of problem?
Use the table below to jump to the right page based on what you are seeing.
| Symptom | Go to |
|---|---|
| A node shows offline or won't reconnect | Node Won't Connect |
| A connector test fails or sync returns errors | Connector Sync Failures |
| Desktop session won't start, drops, or shows a black screen | Desktop Session Issues |
| Terminal session won't start or stalls | Terminal Session Issues |
| Can't log in, session expires immediately, SSO problems | Login and Authentication Issues |
| Console feels slow, pages take a long time to load | Slow UI and Performance |
If the problem spans multiple categories -- for example, all nodes are offline and you also cannot log in -- start with the hub health checks above. A hub-level issue often explains multiple symptoms at once.
Tip: When you open a support thread or file an issue, include the output of
/healthz,docker compose ps, and the relevant hub logs. This saves a round trip of back-and-forth.