[Bug]: The engine crashed after a while

[bbarnabas]

JASP Version

0.19.1

Commit ID

No response

JASP Module

T-Tests

What analysis are you seeing the problem on?

Independent Samples T-test, one-sided

What OS are you seeing the problem on?

Windows 11

Bug Description

While changing between one-sided (left) and one-sided (right), the engine crashes...

hf1.zip

Expected Behaviour

Not crashing.

Steps to Reproduce

I have checked in the following:

1. Student, Welch
2. Normality, Equality of Variances (Levene), Q-Q plot
3. Descriptives
4. Descriptive plot
5. Then changing to one-tailed left and right couple of times.
   However, the problem is intermittent, that is one possible sequence, and it does not trigger it every time. real sequence is unknown.

Log (if any)

No log, but turned it on.

More Debug Information

0.19.1

Final Checklist

• I have included a screenshot showcasing the issue, if possible.
• I have included a JASP file (zipped) or data file that causes the crash/bug, if applicable.
• I have accurately described the bug, and steps to reproduce it.

[bbarnabas]

Sometimes just crashes when leaving alone JASP for 10 min, then going back and by starting a new test (e.g. Paired Samples T-test).

[bbarnabas]

LOG:
JASP 2024-11-22 log.zip

[tomtomme]

@bbarnabas
thx for the report.
I cannot recreate the error on my machine. Have you been able to find an exact sequence to trigger the error?
Have you also maybe updated the OS and restarted the system to fix this?

@boutinb
any idea? The logs do not tell me much and I cannot test on win11.
The provided data file looks fine to me.

[boutinb]

What is interesting in the log is this:

00:49:55.287: AnalysisForm::formCompletedHandler for 000002B152E3CD50 called.
00:49:55.288: AnalysisForm::setAnalysisUp() for 000002B152E3CD50
00:49:55.308: Registering engine #1 for module 'jaspTTests'
00:49:55.330: Engine #1 had nothing to do for so long it has decided to shutdown.
00:49:55.330: Engine #1 shutEngineDown()
00:49:55.330: Engine #1 stopEngine(), _stopRequested will be set
00:49:55.330: informing engine #1 that it ought to stop
00:49:55.331: EngineRepresentation::processEngineStoppedReply() for engine #1
00:49:55.331: Engine #1 shutEngineDown()
00:49:55.331: Engine #1 stopEngine(), _stopRequested will be set
00:49:55.331: ~EngineRepresentation() Engine #1
00:49:55.336: R_HOME set to C:/PROGRA~2/JASP/R
00:49:55.336: Windows PATH was changed to: 'C:/Program Files (x86)/JASP;C:/Program Files (x86)/JASP/R/bin;C:/Program Files (x86)/JASP/R/bin/x64'
00:49:55.535: Setting new engine process to engineRepresentation Engine #0
00:49:55.535: Not enough idle engines! Need 1 for analysis, one will be idle soon...
00:49:55.535: Registering engine #0 for module 'jaspTTests'
00:49:55.541: R_HOME set to C:/PROGRA~2/JASP/R
00:49:55.541: Windows PATH was changed to: 'C:/Program Files (x86)/JASP;C:/Program Files (x86)/JASP/R/bin;C:/Program Files (x86)/JASP/R/bin/x64'
00:49:55.549: Setting new engine process to engineRepresentation Engine #1
00:49:55.553: Finding/constructing all relevant shared memory objects again.
00:49:55.554: Finding/constructing communication strings
00:49:55.554: Creating JASP-IPC-29524_1-md1
00:49:55.554: Creating JASP-IPC-29524_1-sd1
00:49:55.555: informing engine #0 that it may resume.
00:49:55.557: informing engine #1 that it may resume.
00:49:59.251: Engine # 1 finished while in state 'resuming' crashing and with exitCode -1073740791!
00:49:59.251: EngineRepresentation::handleEngineCrash():
**Engine # 1** process *crashed* and it's state is *resuming*.

So apparently the engine is asked to stop because he is doing nothing, but just some milliseconds after, it is asked to resume.
This cannot be only a really bad luck: apparently this has happened several times. @JorisGoosen can you have a look at this?

[bbarnabas]

It matches the problem that I have encountered: the crash happened several times when I left the opened db in JASP for like 10 min and did not interact with it (I was doing sth else on the laptop), then I resumed to the program, wanted to start some analysis, then it crashed.

[JorisGoosen]

Hmm, Ill have a look and see if I can reproduce it somewhere and somehow