error using parallel pool (parpool)

Hi! im trying to load a simulation file in a simulator app that i have (it run with matlab runtime). suddenly the app got an error (it repeat in multiple computers).

the 'log' text when loading the simulation file in the app is:

Starting parallel pool (parpool) using the 'local' profile ...

Starting parallel pool (parpool) using the 'local' profile ...

Error using parpool (line 145)

Parallel pool failed to start with the following error.

Error in appdesigner.internal.service.AppManagementService/tryCallback (line 347)

Error in appdesigner.internal.service.AppManagementService>@(~,~)obj.tryCallback(app,startupFcn,false,[]) (line 156)

Caused by:

Error using parallel.internal.pool.InteractiveClient>iThrowWithCause (line 670)

Failed to locate and destroy old interactive jobs.

Error using parallel.Cluster/findJob (line 74)

Unknown type: concurrentconcurrent.

Error using matlab.apps.internal.GraphicsCallbackProxy/runCallbackFunction (line 23)

Error while evaluating GraphicsCallbackProxy CallbackFcn.

i followed the instructions in this link but it doesnt work. i also enter the command:

 distcomp.feature( 'LocalUseMpiexec', false )

in the command window and it still not working.

the app is running only in a specific computers (clasified) so i cant share it

Note: I did a validation to my local cluster and the only unusual note is that the 'Job' ran with 6 workers.

Answer 1

[Gayatri]

Hi Roee,

The error message related to Failed to locate and destroy old interactive jobs and Unknown type: concurrentconcurrent suggests there might be a problem with the management of parallel jobs or a misconfiguration in your setup.

Here are a few steps you can try to diagnose and potentially resolve the issue:

• Clean Up Parallel Pool Environment: Before starting your application, ensure that any existing parallel pools are shut down and that there are no jobs that might be causing conflicts.

>> delete(gcp('nocreate')); % This command shuts down any existing parallel pool

• Check and Reset Parallel Configuration: Inspect your current parallel configuration and reset it to the default settings.

parallel.defaultClusterProfile('local'); % Reset the default profile to 'local'

• Validate Parallel Pool Setup: After cleaning up, try to manually start a parallel pool to see if the issue persists outside of your application environment. This can help in identifying if the issue is with the MATLAB parallel setup or something specific to your application.

>> parpool; % This command attempts to start a parallel pool using the default configuration

I hope it helps!

[Roee]

Hi Gayatri,

Thanks for helping out!

Well, actually the issue never persisted outside my application, which runs on matlab runtime. Maybe I should've pointed it out in my first post. To make sure I followed your steps anyway, and the issue persists on my application.

Any ideas regarding the difference between parallel jobs management when running MATLAB versus when running a MATLAB-runtime based application? I think this might be the key here.

Thanks again!

Roee