ML Process died (exit code 139)

Highlighted
UserBird Dataiker
Dataiker
ML Process died (exit code 139)



 



From the log:



python(5787,0x70000fcc5000) malloc: *** error for object 0x7ff8fe7317e0: incorrect checksum for freed object - object was probably modified after being freed.



[2018/01/24-14:07:02.303] [Exec-158] [INFO] [dku.utils]  - *** set a breakpoint in malloc_error_break to debug



[2018/01/24-14:07:02.304] [Kernel-159-monitor-159] [INFO] [dku.kernels]  - Process done with code 134



[2018/01/24-14:07:02.408] [MRT-153] [ERROR] [dku.analysis.prediction]  - Processing failed



com.dataiku.dip.exceptions.ProcessDiedException: ML process died (exit code: 134)



at com.dataiku.dip.exceptions.ProcessDiedException.getExceptionOnProcessDeath(ProcessDiedException.java:46)



at com.dataiku.dip.kernels.DSSKernelBase.getExceptionOnProcessDeath(DSSKernelBase.java:129)



at com.dataiku.dip.analysis.coreservices.AnalysisMLKernel.executeCommand(AnalysisMLKernel.java:105)



at com.dataiku.dip.analysis.ml.prediction.PyRegularNoSavePredictionHandler$TrainAdditionalThread.process(PyRegularNoSavePredictionHandler.java:118)



at com.dataiku.dip.analysis.ml.shared.PRNSTrainThread.run(PRNSTrainThread.java:66)



 



****



 



Is this a caching issue?

0 Kudos
1 Reply
Clément_Stenac Dataiker
Dataiker
Re: ML Process died (exit code 139)
Hi,

This looks like a memory corrumption bug in one of the underlying numerical computation libraries (numpy, pandas, blas,....). Is it reproducible ? Reproducible with other algorithms on this dataset ? Could you share details about your setup ? Are you at a liberty to share this dataset ?
0 Kudos
Labels (1)