We recently enabled a second node in a Process Automation Cluster, and intermittently the new node fails to successfully run Script Operators (powershell). When this occurs StartTime and EndTime are usually 1 second apart and the scriptOutput variable is null/empty. I suspect the script is not running at all as I don't think any of our scripts can run in that short amount of time. This occurs across multiple workflows. I have reset the operator after the issue occurs and it may succeed on a retry, but after a few more operators the behavior starts again.
For example, in one instance one script operator exhibited the behavior (the start and stop times were actually identical) and two operators later (the next script operator in the workflow) the operator was executed properly.
I did see an article on some invalid settings in OasisConfig.properties, but the settings on this server match the suggested vaules.
#Specifies whether the user selected an option to change the PowerShell run policy during installation.
#Defines the PowerShell path on the host computer.
The problem is happening in a clustered environment. Using Identify which server an operator is running on in an ITPAM cluster we are able to identify that the script only has these problems on one of the two nodes/servers in the cluster. Furthermore, we can see that one the problematic node/server - the problem is intermittent.
Release : 4.3
Component : Process Automation
An onsite colleague/engineer suggested a security product might be causing this issue. We had the admin stop the product, and at that point was unable to recreate the issue. Focus is on the security product. Since stopping and starting (to see if it can be reproduced) the problem did not reoccur for a considerable amount of time. It did eventually reoccur - though it was while 20 processes were waiting to run at one time. Otherwise, daily operations have not been impacted like they were previously.