SystemLink Analysis Automation Execution Exceeded Maximum Duration

Issue Details

• On my SystemLink Server, I have configured an Analysis Automation Instance. Whenever it's triggered I receive an error stating execution exceeded maximum duration.
• I have created an Analysis Automation Procedure (ANP file) with a script. Whenever the script executes from SystemLink, I see a timeout like below.

Execution exceeded maximum duration (60000 ms) while executing event:
On_Run_AnalysisProcedure


 

Solution

This error occurs because the Analysis Automation Procedure (ANP file) script takes longer to execute than the specified timeout allows. It can be resolved by increasing the timeout or reducing the execution time of the script. Work through the steps below to resolve this issue:
 

• Increase the global Analysis Server timeout.
  1. In the SystemLink web interface, select Measurement Data Analysis >> Analysis Automation.
  2. Click Instance to view a list of instances.
  3. Select the Settings tab.
  4. Increase the Global Timeout [Min] setting.
  5. When prompted, click Apply to save changes.

 

• Increase the timeout of the individual ANP scripts.
  1. Open DIAdem and select the SCRIPT panel.
  2. On the top menu bar, select Settings >> SystemLink TDM >> Analysis Automation Procedure.
  3. Open the ANP that is timing out.
  4. Select the Analysis Script tab.
  5. Increase the Timeout [Min] setting and save the ANP.
  6. Update the Analysis Automation Instance on SystemLink with the new ANP file.

 

• Maximise the utilization of Compute Nodes and Parallel Requests to Compute Nodes.
  1. Open the NI SystemLink TDM Configuration application.
  2. Select the Parallelization tab.
  3. Increase the value of Number of compute node instances. This setting relates to the number of Windows processes that will execute work.
     • The maximum value is 64, and does not directly correspond to the number of CPU cores on the server.
     • It may be necessary to benchmark several values since some PCs will see diminishing (or even reduced) return beyond 16.

 

4. In the SystemLink web interface, select Measurement Data Analysis >> Analysis Automation.
5. Click Instances to view a list of instances.
6. Navigate to the Settings tab.
7. Configure the Number of Parallel Requests to the Compute Nodes option. This setting relates to the maximum number of jobs that can wait in the queue for a Compute Node.
   • The maximum value is 64, and does not directly correspond to the number of CPU cores on the server.
   • For maximum efficiency, the total sum of Number of Parallel Requests to the Compute Nodes across Analysis Automation and Data Preparation Instances should be equal to or less than the number of Compute Nodes.