# Troubleshoot the txAdmin recipe

Use this flow when recipe deployment stops, a downloaded resource is missing, or the first boot
cannot connect to the database. Do not rerun destructive database steps over a live schema.

## Evidence to collect

- The failing txAdmin step and its first error, not only the final summary
- Recipe URL/commit, artifact version, OS, and txAdmin version
- Whether database creation, repository download, or server start failed
- A redacted copy of connection settings with passwords and license keys removed

## Safe diagnostic

1. Compare the selected recipe with current `qbcore.yaml` upstream.
2. Confirm the host can reach GitHub and the configured database without printing credentials.
3. Inspect the new deployment directory and database before retrying.
4. Retry in a fresh temporary profile/database so partial state cannot be mistaken for success.

For general txAdmin failures involving downloads, file validation, or database tasks, see the [independent recipe troubleshooting guide](https://txadmin.org/docs/troubleshooting/recipes).

## Fix

Correct the first failing prerequisite, then deploy a fresh profile from the official recipe. If a
repository or recipe step changed upstream, pin and record the reviewed commit before production.

## Rollback

Stop the failed profile. Restore the pre-deployment database backup if an existing schema was
touched, then remove only the explicitly identified failed profile after confirming its path.

## Sources

- [Official txAdmin recipe](https://github.com/qbcore-fivem/txAdminRecipe)
- [Public Cfx.re recipe failure pattern](https://forum.cfx.re/t/fivem-setup-run-recipe-for-qbcore-error/5226697)
