Replay does not work with Preview Serverless (AI scaling)

[cjsekl]

Running Replay on a Serverless workgroup on the AI Scaling preview track fails with:

Traceback (most recent call last):
  File "/home/ec2-user/redshift-test-drive/common/util.py", line 93, in cluster_dict
    cluster["base_rpu"] = response["workgroup"]["baseCapacity"]
KeyError: 'baseCapacity'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/ec2-user/redshift-test-drive/core/replay/replay.py", line 237, in <module>
    main()
  File "/home/ec2-user/redshift-test-drive/core/replay/replay.py", line 42, in main
    cluster = cluster_dict(config["target_cluster_endpoint"], is_serverless_endpoint)
  File "/home/ec2-user/redshift-test-drive/common/util.py", line 95, in cluster_dict
    if e.response["Error"]["Code"] == "ResourceNotFoundException":
AttributeError: 'KeyError' object has no attribute 'response'

I'm guessing because these workgroups don't have base capacity. It looks like the exception block is meant to handle it gracefully but doesn't cover this case.

Worked around it by setting cluster["base_rpu"] = "N/A"

[cjsekl]

Ignore me, works with later releases