Major Description
While running cluster tests with failing jobs I noticed that the JobManager log file does neither contain the root cause of the failure nor a stack trace of it. The error was not sufficient resources available for scheduling.
For the purpose of debugging, it is crucial to have this information in the log file.