There is an intermittent recurrig problem where the ITPAM workflow gets stuck in a "running" or "suspended" status and cannot be aborted via the UI. This cannot be reproduced at will, but sometimes works and other times does not. When this fails it ALWAYS gets stuck on the operator that is reading a JSON file. This never happened before upgrading to ITPAM 4.4 The problem appears to be happening in a script that reads and processes a json file.
The current investigation points to a structural persistence failure where Orchestrator nodes fail to maintain unique RefID sequences across the cluster. The proposed resolution path involves: