Forum Discussion

Alleyn's avatar
Alleyn
Contributor
3 years ago

LoadUI Not all scenarios running

I have 12 scenarios created for a LoadUI test. Each one has just one test case in each with a wait time of 1ms. As a proof of concept before kicking off my run I set all scenarios to have 1 VU each. However, ten minutes in and only two scenarios are running. If I stop the run and kick it off again I could get just 1 scenario running. The run has Equal Load for all scenarios - why the randomness?

  • Hi Alleyn ,

    I tried replicating your issue by creating a simple load test with 12 scenarios. Executed with 1 virtual user in each with a 2 second wait time after each VU. My results looked something like this. 


    You can see that every 2 seconds or so the TPS count hits 12 TPS and VUs hold constant at 12 which indicate all scenarios are executed. I also set up logging conditions and statistic groups to confirm that each scenario executed.  

    If you could provide more information on your setup that would be helpful for troubleshooting what is causing your issue. There's also logging options at the bottom of your ReadyAPI UI that includes a log for performance tests. This might have some indication on why your scenarios aren't all executing. 

  • Hi Alleyn ,

    I tried replicating your issue by creating a simple load test with 12 scenarios. Executed with 1 virtual user in each with a 2 second wait time after each VU. My results looked something like this. 


    You can see that every 2 seconds or so the TPS count hits 12 TPS and VUs hold constant at 12 which indicate all scenarios are executed. I also set up logging conditions and statistic groups to confirm that each scenario executed.  

    If you could provide more information on your setup that would be helpful for troubleshooting what is causing your issue. There's also logging options at the bottom of your ReadyAPI UI that includes a log for performance tests. This might have some indication on why your scenarios aren't all executing. 

  • Hi SarahS 

     

    Thanks for getting back to me. I did some further analysis just now and it was actually an erroneous DB search that had the wrong IP Database selection so couldn't pull back the data. That must have impacted the only run then because when I fixed that the run was successful. 

     

    Btw is there any way I can include database source steps from the end results? It doesn't make sense to have a Database query taking 30 seconds to pull back test data being report as a badly performing result in the LoadUI report when it has nothing to do with the API call itself. Happy to open a new  discussion on this one!

     

    Thanks again for your previous input!

    • SarahS's avatar
      SarahS
      Staff

      The reports typically include information about the targets/test cases you add in there, including information on the test steps. As a quick example I ran a simple load test with a data source in it. The resulting report shows metrics for that data source. 

      Other things that you might try doing to retrieve more information about your datasource test step would be to set up statistic groups or logging conditions. 

      Those are just a few ideas that come to mind on getting information from load tests. It might also be a good idea to open this up as a new thread to get more feedback from the community in general.