Capsules
Last updated
Was this helpful?
Last updated
Was this helpful?
This section covers:
To add a capsule to the pipeline, drag the capsule from the Add Capsule menu and drop it into the pipeline building area.
If the capsule is not ready to be used in a pipeline a hazard symbol appears beside it. An up-to-date reproducible run is required before a capsule can be used in a pipeline, and all changes must be committed in the capsule for those changes to be reflected in the pipeline. More information on committing changes in a capsule can be found in .
When a capsule is dragged and dropped with a released version into the pipeline building area, you can select which version to use.
The version can be changed at any time in the capsule's settings.
The version used by the pipeline will remain unchanged if a new version of the capsule is released.
Hover over the capsule and click the garbage can to remove it from the pipeline.
The replace feature can be used to replace capsules while maintaining all connections and mappings.
When a capsule that uses secrets is added to a pipeline, the same secrets that are selected in the capsule's environment editor will be used by the pipeline. If the required secrets have not been selected, users will receive an error when trying to run the pipeline. The error message provides links to all the capsules that require secrets, making it easy for users to identify which capsules need to be updated.
For example, if the "Accessing Redshift database" capsule requires secrets and they have not been selected in the capsule, users will receive the following error when trying to run the pipeline:
Once the capsule is ready to be used in a pipeline, it is best practice to release the capsule. This allows the released version to be used in the pipeline which ensures any subsequent changes to the capsule do not affect the pipeline. See .
In Add Capsules, capsules with a released version are indicated by the released icon .
AWS Cloud Credential secrets that are used to access external data assets will not be used by the pipeline. A custom IAM role is required to use external data assets in a pipeline. See for more information.
