If I clone an existing job without making any changes, I am able to reconfigure the compute successfully. If I remove or add a spark environment variable to test modifications, such as using secrets for example, and I confirm the changes to the job, I lose the ability to configure the cluster compute, which essentially defeats the purpose of cloning the job in this case.
I've done this in previous iterations of the UI without issue but this is the first time in recent history. Is this expected behavior?
