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

# E-SCHED-001

> Retired: doctor no longer reads the live machines route to see whether a schedule caller exists.

```text retired theme={null}
E-SCHED-001 — retired, doctor no longer emits this code
```

`error_code: E-SCHED-001` · `status: retired`

## What you're seeing

Nothing: `npx vendo doctor` cannot print this code anymore. You are reading
it from an older run's output, or a saved report.

## Why

This warning fired when apps declared `vendo.json` schedules and no caller
was authorized to trigger them. Doctor read that from the live
`/doctor/machines` route. It reads files only now, so it never sees the
running app's schedules.

## The fix

The gap is still real, and you check it yourself now. If any app's
`vendo.json` declares a schedule, something has to be able to call
`POST /api/vendo/tick` or the schedules never fire. With a Cloud key it already
can: the deployment derives the tick secret from `VENDO_API_KEY` and registers
itself at boot, and Cloud's heartbeat knocks once a minute. Without one, set the
secret in the server environment yourself.

```bash .env.local theme={null}
VENDO_TICK_SECRET=<a random secret>
```

Point an external cron (Vercel cron, a GitHub Actions schedule, plain
crontab) at the tick route with that secret as a bearer token:

```bash {2} theme={null}
curl -X POST https://your-app.com/api/vendo/tick \
  -H "Authorization: Bearer $VENDO_TICK_SECRET"
```

Every hit fires due schedules exactly once for that window; a double hit
from an overlapping cron is a safe no-op.

## Related errors

* [E-CFG-001](/production/troubleshooting/e-cfg-001) — a missing `.vendo/` file, still checked on disk
* [E-LIVE-004](/production/troubleshooting/e-live-004) — the execution venue scheduled apps run on, also retired