Recipe deployment problems
Identify the first failing task and avoid damaging an existing server.
Start with the first failed task
Recipe tasks run sequentially and stop on error. Record the action name, source URL or file path, and full error. Re-running the deployment before understanding the failure can overwrite files or repeat database operations.
Download failures
Check whether the repository, branch/tag/commit, and file URL exist. A private repository needs the appropriate access. GitHub rate limits and transient network errors can also stop a deployment. The upstream docs recommend explicit references for recipes with many download_github actions.
Do not substitute an unknown mirror for a failed download without checking what it contains.
Database failures
Confirm the host, port, database name, credentials, and permissions. In a container, localhost refers to that container, not a separate database container. Verify network reachability and the recipe’s expected database engine.
connect_database must run before query_database. Inspect the failing SQL and preserve the database before retrying any destructive query.
File and validation failures
Check that the target folder is writable and paths stay within the recipe’s deployment directory. At completion, the deployer expects a server.cfg file and a resources folder. Inspect the download/extraction paths if those are missing.
Retry safely
Use an isolated empty folder and test database when reproducing a recipe issue. If a previous attempt partially deployed files, understand that state before starting again. Never redeploy a troubleshooting recipe over your production resource directory.
See Recipe files reference for supported actions and options.
Source: txAdmin upstream documentation.