Databricks Asset Bundle Workflow Redeployment Issue

[Sikki]

Hello All,

In my Databricks workflows, I have three tasks configured, with the final task set to run only if the condition "ALL_DONE" is met. During the first deployment, I observed that the dependency "ALL_DONE" was correctly assigned to the last task. However, upon subsequent redeployments, I noticed that "ALL_DONE" was also being assigned to other tasks. Upon further investigation, it appears that the naming of the task keys may be causing this discrepancy. This hypothesis was confirmed when I tested with different task key names, such as task-A, task-B, and task-C, and found that the workflow configuration was correctly maintained during redeployment as well.
Thanks,Sikki

[Sikki]

Hi, I've attached two snapshots, one from the initial deployment and another from the second redeployment, where no changes were made to the code.

[Yeshwanth]

Hi @Sikki Good day!

There was an issue and it was fixed recently. Could you please confirm if you are still facing the issue?

Best regards,

[Sikki]

The issue hasn't been resolved yet. Should I update my Databricks CLI?

[Sikki]

The issue hasn't been resolved yet. Should I update my Databricks CLI to generate new bundle_config_schema.json.

 

[Yeshwanth]

@Sikki are you using the latest version of the CLI? v.28?

[Sikki]

Hi @Yeshwanth I am using latest CLI version and using Azure DevOps to deploy the workflow.

[Sikki]

After updating my CLI, I successfully deployed the job from Databricks CLI and it is functioning correctly. However, when attempting to deploy the same job using Azure DevOps, I encounter the same issue.

[Yeshwanth]

@Sikki thanks for confirming.

For Azure DevOps please check the version of Databricks CLI installed.