Your pipeline is executed in a short-living Docker container which is dropped when the pipeline finishes. So if your artifacts are written to the container's filesystem, those will be lost.
We were a similar position like you, we needed to save the results of the pipeline and make downloadable by human users.
We did this:
- During the pipeline execution, after the artifacts are produced, we create Bitbucket downloads from those using the REST API. It is super-easy. We use a timestamp as suffix of the artifact's filename.
- Then if someone needs an artifact, he can go and download it any time.
- (Because we produce a lot of these artifacts, the pipeline will even delete those downloads that are older than N days, so we always have the artifacts produces in the last N days.)
It works fine and it has been stable for many years.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.