i know the simple answer to this question is to go into the project properties (current and/or default) and set the "error handling", "on error" and "on warning" drop downs to the "Continue running" value. I've done that (see attached image) and my test cases still periodically come to a grinding halt and I have to manually click the green triangle play button to continue. Is there some other settng somewhere else that i can check? I would note that this only seems to happen on one machine where I sometimes run my scripts. it doesn't occur on my own workstation but i've checked the properties page on it and they are set the same. it's got the same version of test complete as my workstation. I've searched the FAQ's and there's a few posts about using properties to stop the execution but nothing i could find about how to prevent a run from stopping when you don't want it to.
Solved! Go to Solution.
I think i have found the solution to this in a different post that relates to the same issue. The post is entitled "TC always paused on error". In it, Stephen suggests "Can you check under Tools >> Options >> Engines. It may be that the 'Pause test execution on posting an error option' is ticked." On my workstation it wasn't ticked but on the workstation that was giving me the problem, it was ticked. I'm running a test on that machine now to see if it does or does not stop on an error but I'm pretty confident this will be the resolution.
Shouldn't there be some sort of reference on the playback screen (where all those 'continue running' on errors and warnings are located) to check this option? would have saved me several days of research if there was.
Thanks Stephen. How would I accept his solution on the other post as the solution to this post?
The best way is to figure out what's causing the delay. What test is running when this happens? What part of the code is being executed?
Is it pausing? Or is it fully halted? Is there a dialog that comes up when it's halted? Do you have a break point dropped in somewhere inadvertently?
Keep in mind, you're looking in the project properties... but there is a nother location in the project that makes these settings. In the Test Items panel, there's an "On Error" and "On Exception" column that allows you to specify settings on a Test Item basis.
More information would be helpful.
Hi @GradyJr , please give tristaanogre and Marsha_R a little more information to work with and the Community will try to solve this for you!
there are no break points and no "test complete" dialog box. it is a pause as i am able to manually click the green play/continue button and proceed from that point on. typically it seems to happen when the application being tested pops up an exception box for which there hasn't been a "click ok" recorded but that's not the case every time. some times it just pauses and it's only because the playback failed a property checkpoint. Something on the screen isn't equal to the case sensitive value that it was during recording. i checked the test items tab and everything there is set to "continue" in the event of an error, warning, etc. My manager suggested that there might be some sort of timeout property that isn't set correctly. could that be the issue?
i've automated six project runs for overnight processing. this issue consistently occurs in two of them every night i run it on this particular machine. Please note that i can run these exact same projects on my own personal workstation and they do not ever stop/pause. it is only on this one machine and I am physically logging on to it (not remoting into it from my desktop).
Did you copy the project directly from your machine to the one that pauses or did some/all of it come from a version control tool? If it wasn't a direct copy, try that.
Are the versions of TC exactly the same on the two machines?
i've attached a simple diagram of my environment which shows the version of test complete installed on both machines. Both my workstation and the remote workstation have access to my personal Test Complete project suite as well as the master test complete project suite to which I have staged/committed/pushed changes from my personal working copy. When i have my task scheduler use either copy of the projec t suite it will run from start to finish without stopping. failures and warnings are identified in the log with red or yellow icons but they do not stop the execution. When the same task scheduler set up runs either project suite on the remote workstation ANY error it encounters pauses the execution until the Continue button (green play triangle) is clicked. I've also attached an image of what the log looks like at one particular point where it is pausing. as you can see it simply failed a property checkpoint but wasn't going anywhere until i manually continued it. hope this helps to clarify.
If you log on to the remote workstation manually and run the test without the scheduler, does it work or does it pause?
it still pauses on any error. i started the project run just as you requested without the use of Task Scheduler and then i walked away and came back about an hour later. Test Complete was sitting idle with a pause indicator on the line that received the error. Rather than click the Play button to continue I clicked the Stop button to interrupt the test execution. attached is the screen shot of the log showing the error which is nothing more than a failed property checkpoint. I then clicked on the Project in the Workspace, selected the Properties tab at the bottom and clicked on Playback. as you can see from the second attachment, all error handling is set to "Continue Running". What else do i need to try to make this so that it will log the errors but keep running?