The launcher.exe process crashed.
I have the problem that every now and then, when I let my test run, I get the error message "The launcher.exe process crashed."
The wierd thing about this is that it does not happen every time, only every 10-15th time (or something like that). And also the test doesn't crash, it just continues like normal and while the test is running, you don't even notice anything going wrong, just when I look at the Log afterward, i see that this message appeard.
Below I posted a screenshot of the log (Had to black out some information but those were only Messages)
As you can see the test just contiues like normal, as if nothing happened.
Solved! Go to Solution.
Not sure how much this helps you, but our desktop application crashes occasionally, so we utilize the General Events to interrogate the Log Error message and then stop the current running tests when we find the crash message. You could do something like below. This code assumes that you use Test Cases within the Execution Plan, so you might need to modify it for you specific needs.
Information for setting up events can be found here: Creating Event Handlers for TestComplete Events | TestComplete Documentation (smartbear.com)
function GeneralEvents_OnLogError(Sender, LogParams)
// message to look for…
var messageToFind = "The launcher.exe process crashed.";
// LogParams.MessageText will be the error text
if (aqString.Find(LogParams.MessageText, messageToFind, 0, false) >= 0 && Project.TestItems != null && Project.TestItems.ItemCount > 0)
// a crash might have occured...log a message
Log.Message("A crash might have occurred...stopping the current test: " + Project.TestItems.Current.Name);
// stop the current test only...
Those are the Details provided in the "Details" Tab on the test step that failed.
Exception code: 0x80000003. Process identifier (PID): 3032. Dump file: launcher_ue_7187f00f.dmp.
Microsoft (R) Windows Debugger Version 10.0.22473.1005 AMD64 Copyright (c) Microsoft Corporation. All rights reserved. Loading Dump File [C:\coyote\workspace\Platforms\WebUI\Tests\Git_Update\Coyote\Log\03_01_2022_08_50_37_301\launcher_ue_7187f00f.dmp] User Mini Dump File: Only registers, stack and portions of memory are available Symbol search path is: srv* Executable search path is: Windows 10 Version 19042 MP (12 procs) Free x64 Product: WinNt, suite: SingleUserTS Edition build lab: 19041.1.amd64fre.vb_release.191206-1406 Machine Name: Debug session time: Mon Jan 3 09:41:54.000 2022 (UTC + 1:00) System Uptime: not available Process Uptime: 0 days 0:00:01.000 .................................................. This dump file has a breakpoint exception stored in it. The stored exception information can be accessed via .ecxr. For analysis of this file, run !analyze -v launcher!Launch+0x640487: 00007ff7`4c37bad7 cc int 3
First of all, thanks for your reply.
As far as I can see, your function just looks for a crash that is specifically the one that I get and the prints out the name of the test that failed. But I am not sure how this could help me, since I already know which test failed.
Maybe to make it a little bit clearer for you how I run these tests:
I have a big execution plan with many tests in them but I only run 8 of them by highlighting them and "run focused item". So only 8 Tests run. These 8 Tests are all the same Keyword tests just with 3 Parameters (Booleans) that change every time (2³ -> so 8 Tests). This Keyword test that I run only runs 1 Python script. So in the test results I can see which of them failed. But the weird part is that all of them run except for 1 or 2. But the ones that fail aren't the same ones everytime, its sort of random. And I don't really know what causes this issue.
The issue was caused by having Opera open in the Background. The "launcher.exe" that has crashed is launcher of Opera. For some reason having Opera open in the Background while this specific Test is running, there are some interferences. Just closing didn't really help me, so I just renamed the "launcher.exe" to something else. There might be some issues with Opera in the future for me, but as long as my test is running, I am happy.
Hope this helped if somebody comes across this.
Thank you! I reviewed the .dmp file created on the error and it seems that the "launcher.exe" is the launcher of Opera (my tests where running in Chrome though) and somehow having Opera open in the background messed with my test. Renaming the launcher.exe solved this issue, though I might run into issues with using Opera in the Future. This doesn't really matter, because our tests are running on sperate machnienes, that don't have Opera installed.
Thanks for your help.