Scheduled Runs not running latest version of test

2 Answers 37 Views
Remote Execution Test Execution Test Recording
Adrian
Top achievements
Rank 2
Iron
Iron
Iron
Adrian asked on 26 Feb 2024, 11:34 PM

Hi,

I quite often find that scheduled tests are not picking up changes I have made to the test or Elements within that test.

e.g. if I change the properties of an element that can't be found and run it individually the test picks up those changes and passes but when run in a scheduled list it isn't picking up the changes.

Why does this happen and what can I do to fix it.  Very annoying.

Cheers

2 Answers, 1 is accepted

Sort by
0
Steven
Top achievements
Rank 2
Iron
answered on 28 Feb 2024, 04:38 PM

I had an issue that was similar yet not quite the same with running Scheduled Tests, so here are a few suggestions:

  1. Before you run your test make sure that you verify the files where changes were made (see changes.png). I have on some occasions inadvertently unchecked a box by a file.
  2. If you are scheduling a test with an embedded Test Studio for APIs Test, ensure you right click on the project in the Test Studio for APIs Test and select "Compile (F6)". Sometimes if it's not recompiled changes don't flow through.
  3. After several Test Studio updates over time, I ended up with an out of date .dll file that wrecked havoc on my ability to run Scheduled Tests. A fresh install where you delete the "Test Studio Folder" from the "C:\Program Files (x86)\Progress" directory was the only fix.

Anyway, that's it for me - have fun testing!

SSq

Elena
Telerik team
commented on 29 Feb 2024, 10:10 AM

Hi SSq, 

Thanks for adding your advice on the topic! I appreciate it!

Best,
Elena

0
Elena
Telerik team
answered on 29 Feb 2024, 10:09 AM

Hello Adrian,

Do you upload the changes applied in the project to the Storage database?

Please, note that this isn't an automatic action and you need to keep the project version in the Storage DB up to date with the desired copy of the project. In long term this allows you to work on the project in different branches but keep the scheduled jobs running with a stable project version. 

I hope this information is useful for you.

Regards,
Elena
Progress Telerik

Virtual Classroom is the free self-paced technical training portal that gets you up to speed with Telerik and Kendo UI products including Telerik Test Studio! Check it out at https://learn.telerik.com/.
Adrian
Top achievements
Rank 2
Iron
Iron
Iron
commented on 01 Mar 2024, 03:28 AM

Hi Elena,

We run everything off the server here.  i.e. creating tests and running tests are all done directly on the same server that executes the tests (Run locally).

Do you still need a storage DB in this instance?

Thanks

Adrian

Elena
Telerik team
commented on 01 Mar 2024, 01:15 PM

Hi Adrian, 

Whether the Test Studio Services are in use depends on the setup you have on your end. The Test Studio Scheduling configuration can be enabled also on a single machine and is the recommended setup when you intend to schedule test list runs.

If you share an image from the Machines status view I'll know what is the setup you use and if uploading the changes is a must for the scheduled jobs. 

Please, have in mind that this is a public thread. So, I will recommend to submit a new private support ticket if you have any concerns sharing the screenshot. 

Thanks for your cooperation once again.

Regards,
Elena

 

Adrian
Top achievements
Rank 2
Iron
Iron
Iron
commented on 04 Mar 2024, 01:17 AM

 

Thanks

 

Adrian

Elena
Telerik team
commented on 04 Mar 2024, 10:04 AM

Hi Adrian, 

This setup uses the local project copy for the test list execution so I don't expect to see differences in the version of elements and test steps. There's only one thing that comes to my mind - you have applied some changes, but these were not yet saved when the scheduled test run started. 

Other than that I assume you may be seeing differences in the outcome of a test executed as a single test run through the Quick Execute button or when the test run is part of a test list. If this is the case, then I will recommend you to revise the test steps and enhance these with sufficient wait and verification steps, and where necessary execution delays. Check out this blog post which talks about using these kind of steps and how to utilize them best. 

In case you are still not sure what adjustment you may need to apply, send me a copy of the execution log for the successful test run and the step failure details for the failing test list run (you can use the Export button to get a zipped folder with all failure details). These will help me revise the steps and suggest you possible improvements. 

Feel free to correct me if my assumptions are not pointing to the exact issues you experience.

Thank you for your cooperation in advance.

Regards,
Elena

Tags
Remote Execution Test Execution Test Recording
Asked by
Adrian
Top achievements
Rank 2
Iron
Iron
Iron
Answers by
Steven
Top achievements
Rank 2
Iron
Elena
Telerik team
Share this question
or