> ## Documentation Index
> Fetch the complete documentation index at: https://docs.still200.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Troubleshooting

> Resolve common HTTP and scheduled-job integration problems.

## HTTP monitors

### The validation response has `checks: null`

Still200 could not parse a valid check set. Confirm the endpoint is publicly reachable, returns HTTP `200` within five seconds, and responds with JSON matching the [health-check specification](/http-monitoring/health-check-spec).

### A degraded dependency did not alert

This is expected. Degraded results are recorded, but only unhealthy or failed polls count toward the three-failure incident threshold.

## Scheduled jobs

### A ping returns `401 Unauthorized`

Confirm that the `Still200-Ping-Token` header is present and contains the latest token. If the token was regenerated, update the job's secret store because the old token became invalid immediately.

### A start ping returns `409 Conflict`

The monitor may be paused, missing schedule data, or starting more than 60 seconds before its expected run. Review the response's `detail` value for the exact reason.

### Finish or fail returns `409 Conflict`

There is no active run. Each finish or fail ping must follow a successful start ping for the same monitor.

### A job stays `running`

The process did not send a finish or fail ping. Configure a maximum runtime so Still200 can mark abandoned runs as `timed_out`, and make sure every exit path reports a terminal state.

### The job was marked missed but ran later

Send the start ping when the retry begins. Still200 can reopen the missed occurrence as `running`, resolve its missed-run incident, and record it as `late` after a successful finish.