Administer pipelines across instances - Data360_DQ+ - Latest

Exporting definitions

1. Select the Pipelines menu at the top of the page.
2. Select the pipelines or data stages that you want to export.
3. Click the menu button at the top of the Pipelines browser:

   

4. Select Export Definitions.
5. An Export dialog will ask you if you want to reset the "dirty state" for items being exported.

   The "dirty state" of an item indicates whether the item's definition has changed since the last time it was promoted or exported. In the Pipelines view, you will see that some items have an asterisk to the right of their name. This indicates that the item has a "dirty state", meaning that the item's definition has changed since it was last promoted or exported.

   If you choose to reset the dirty state when exporting, the dirty state of the item will be set to false and the asterisk will disappear in the source environment. The asterisk will be displayed again if the data stage is edited after the export.

   Select Yes or No, as required.

   A zip file containing the export definitions will be downloaded.

6. Save the zip file somewhere accessible as you will need this file when importing to your new instance.

What is included in an export?

When you export the definitions of a data stage, you are exporting the behind the scenes rules that define the data stage. Exports do not include any data that is associated with a data stage. For example, if you export a data store and then import it to another instance, the data store will contain all of the fields and settings that you applied to it in its original instance. However, the data store will not contain any data. This means that after importing to a new instance, you will need to reload your data stages with the relevant data.

What should I export?

Although you can export any individual data stage, it is recommended that you export entire pipelines. This is because to properly import a data stage definition, the destination instance must contain the parents of the data stage (its pipeline and path) and any dependencies that the data stage may have.

For example, if you exported an analysis definition and then attempted to import it to another instance, the import would fail because the parent pipeline and path of the analysis do not exist in the new instance and the data stores that the analysis uses do not exist in the new instance. To remedy this, you could bundle the analysis and its data stores into a common pipeline, and then export the entire pipeline definition.

Importing definitions

• Prerequisite: You have exported a data stage and have the zip file that was created during the export.

1. Navigate to your target instance.
2. From the Pipelines view, click the menu button at the top of the Pipelines browser:

   

3. Select Import Definitions.
4. In the Import dialog, browse to the zip file that you want to import.

   Selecting Preserve execution settings in target Analysis and Data Views ensures that execution sizing settings defined on analyses and data views in the instance that you are importing into are not overwritten by the import. This setting does not apply if you are using execution profiles to define execution settings at an environment level, rather than within individual analyses and data views, see Environment execution profile.

   Selecting Reset imported items' dirty states in target environment after import ensures the imported items in the target environment is not-dirty.

5. Click Continue.

   Your data stages will be created in your target instance.

6. Reload your data stages with any relevant data. The import only contains data stage definitions, therefore all newly imported data stages will contain no data and will have zero executions.

Environment groups when importing and exporting

If you have assigned an environment group permission to a data stage, that permission and environment group will be exported along with the data stage definition. When you then go on to import the data stage definition to a new instance, its permissions will already be configured.

Environment group with same name exists on target instance

If the target instance already contains an environment group with the same name as the environment group associated to an imported data stage definition, users within that environment group will automatically have the same permissions to the imported data stage as they did in its source instance. However, keep in mind that the environment group in the target instance may contain different users than it did in the source instance.

Environment group does not exist on target instance

If you import a data stage definition with an environment group association and that environment group does not yet exist in the target instance, the target instance will automatically create that environment group for you. The imported data stage definition will retain the permissions that it had in its source instance, however, the new environment group in the target instance will not contain any users. It is the responsibility of the administrator to add users to the newly created environment group. You can select the same users that had access to the data stage on the source instance, or different users, depending on who is working on the instance, and who needs access to the data stage.

Dashboard controls when importing and exporting

If a dashboard contains controls, they will be exported along with the dashboard and should be present when the dashboard is imported to a new instance. Such controls will be fully functional on the new instance, provided that the dashboard's data view has been exported and imported as well.