Overview
Choose a cache for downloaded software or reusable workspace files and configure its location.
Use caches to reuse inputs and generated files between runs. Choose the guide for the data you want to retain:
| Guide | Use it for |
|---|---|
| Artifacts | Downloaded server and proxy software, Java installations, and resolution metadata; includes offline and refresh settings |
| Snapshots | Selected files or directories produced in a process workspace, such as server libraries |
Configure the cache location
Both use the shared cache root, normally ~/.anvil. Choose another location through
anvil { engine { cacheDirectory } } in Gradle or EngineOptions.cacheDirectory when embedding. See
Directories for the Gradle setup. Keep the same root when
preparing an environment and later running it offline.
Share selected data
For another machine or CI run, copy only the cache data you intend to reuse. The root can also contain private authentication accounts; exclude those account files rather than archiving the entire root. See Authentication.
On GitHub Actions, Anvil Actions does this for you:
whereareiam/anvil-actions/setup@v1 keeps downloaded software, Java installations and workspace snapshots in
the Actions cache and leaves accounts out, and whereareiam/anvil-actions/report@v1 uploads the console logs of
failed scenarios.