We have a job submitted through the Spark Connect API, and running on Serverless Compute.The job got canceled twice and left a total of 14 queries orphan, they are in a weird state because the running time is not increasing, but they are there showin...
