Hot Fix J9K015 for SAS Workload Orchestrator contains the following fixes:

• SAS Workload Orchestrator suspends when you preempt the job. This fix avoids deadlocks.
• The job query returns a missing value as “.” rather than 0.
• Traceback might occur in the High Availability (HA) subsystem when you enable HA trace logging.
• Pending jobs go negative, causing jobs to stop being scheduled.
• Jobs are scheduled on a host with no open slots if the job comes in while a configuration update occurs. As a result, the job uses old host configuration information instead of any new information that you set. For example, host x is set to 0 job slots, and the job comes in while the configuration is updating. You can assign the job to host x if it previously contained available job slots.
• Setting the SAS Workload Orchestrator queue to preempt itself causes SAS Workload Orchestrator to suspend on Wx6. No error message occurs.
• Updating UserDefined resources before the job finishes running causes SAS Workload Orchestrator to log the TKM: Memory Error and crash.
• SAS Workload Orchestrator crashes if you change the grid password while jobs are running.
• The client is suspended and reports a bad error message when you give an invalid value to scheduleThresholds.
• If a host runs a restartable job from a restartable queue and it dies, the job restarts on another host in that queue. When the host comes back up while the job is running on the other host, the host erroneously reports that the job is running on the host that just came back online.
• Additional information will be provided for jobs stuck in a Pending state.
• SAS Workload Orchestrator crashes when removing or adding a service in HA.
• firstEndTime does not work on a SAS Workload Orchestrator jobs call if you submit a job late in the day.
• SAS Workload Orchestrator can crash when you kill the first “master” node and grid rolls over to a “master candidate”.
• You can retrieve the SAS Workload Orchestrator pending reason from the admin Command Line Interface (CLI).