-
The documentation for hashing describes that amongst other things, the following is taken into account:
Would this not defeat the purpose of remote caching? I.E. would a CI runner having pulled latest, not generate different hashes to a developer who happens to have output present on their machine? Or is the idea that CI would exclusively leverage cache from other CI pipelines runs? Alternatively would it be considered best practice when using turbo to ensure all output is removed in development environments prior to a pipeline run? Thanks 👋 |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 6 replies
-
A pipeline doesn't refer to CI, it's kind of like a plan for turborepo. Eg. if I run build make sure to run lint, then build dependencies, then build target. Remote caching allows you to download the cache from local to CI or any other situation. Eg. if I run build on local machine and have remote caching, it will save locally and be put on a remote server. Then in CI it will download the cache, assuming you have the same hash, that you sent to the remote server from your local machine. |
Beta Was this translation helpful? Give feedback.
A pipeline doesn't refer to CI, it's kind of like a plan for turborepo. Eg. if I run build make sure to run lint, then build dependencies, then build target. Remote caching allows you to download the cache from local to CI or any other situation. Eg. if I run build on local machine and have remote caching, it will save locally and be put on a remote server. Then in CI it will download the cache, assuming you have the same hash, that you sent to the remote server from your local machine.