I created a Databricks Asset Bundle and it seems it is successfully being deployed with the databricks bundle deploy
command. When I run the same command again with the same bundle I get the error
The pipeline name 'mypipeline' is already used by another pipeline. This check can be skipped by setting
allow_duplicate_names = true
in the request.
- allow_duplicate_names seems to be a parameter in databricks api but I can't find a corresponding parameter in databricks cli
- Is allowing duplicate names even the right way to go? I think that might create a new pipeline with the same name instead of updating the existing one
The problem was that I was both testing this from my own command line, using my own credentials, and the service principal used by our deployment pipeline. So first I deployed from my machine and then from the deployment pipeline. Seems like the pipelines are connected somehow to the user that created them. So as long as it's the same user deploying everything worked fine.