Selim Raboudi
02/23/2024, 4:51 PMScout
02/23/2024, 4:51 PMdagster_cloud.yaml
file should be placed in the root of your Dagster project. If you have multiple Dagster projects in subdirectories, such as a monorepository, you should add the dagster_cloud.yaml
file to the root of where the Dagster projects are stored, which you have done correctly.
To load your code locations defined by the Python packages pipeline_1
and pipeline_2
, you would specify them in the dagster_cloud.yaml
file using the locations
key. Each location will have a location_name
and a code_source
that specifies how the code location is sourced. Since you are dealing with Python packages, you would use the package_name
parameter within code_source
.
Here is an example of how you might structure your dagster_cloud.yaml
file:
yaml
locations:
- location_name: pipeline_1_location
code_source:
package_name: pipeline_1
working_directory: ./pipelines/pipeline_1
- location_name: pipeline_2_location
code_source:
package_name: pipeline_2
working_directory: ./pipelines/pipeline_2
In this example, location_name
is a user-defined name for the code location that will appear in the Dagster UI. The package_name
is the name of the package containing your Dagster code, and working_directory
is the path of the directory that Dagster should load the code source from.
Please adjust the package_name
and working_directory
as needed to match your actual package names and directory structure. If your pipeline_1
and pipeline_2
directories are structured as Python packages (i.e., they contain an __init__.py
file), then the above configuration should correctly load your code locations. dagster_cloud.yaml reference | Dagster D... button Dagster project files | Dagster Docs button Dagster Databricks Step Launcher package... button Materialization of non-partitioned asset... button Thank you for your feedback!