Error compute kernel failure new Statistics Feature
Hello,
I updated our Dataiku Plateform from 5.X to the 8. version.
So when we try to use the new Statistics feature on a Dataset, I get the following error:
Unexpected EDA kernel failure
com.dataiku.dip.exceptions.CodedException: EDA docker kernel failed to start, caused by: CodedException: EDA docker kernel failed to start, caused by: SecretKernelTimeoutException: Subprocess failed to connect, it probably crashed at startup. Check the logs., caused by: SocketException: Socket closed
[2020/11/25-19:17:01.508] [qtp1528769018-365630] [DEBUG] [dku.tracing] - [ct: 1] Start call: /api/futures/get-update [GET] user=admin [futureId=1EMRVvBF] [2020/11/25-19:17:01.508] [qtp1528769018-365630] [DEBUG] [dku.tracing] - [ct: 1] Done call: /api/futures/get-update [GET] time=1ms user=admin [futureId=1EMRVvBF] thrown=ExecutionException [2020/11/25-19:17:01.508] [qtp1528769018-365630] [ERROR] [dip.controllers] - API call '/dip/api/futures/get-update' failed java.util.concurrent.ExecutionException: com.dataiku.dip.exceptions.CodedException: EDA local kernel failed to start at java.util.concurrent.FutureTask.report(FutureTask.java:122) at java.util.concurrent.FutureTask.get(FutureTask.java:192) at com.dataiku.dip.eda.compute.engine.LazyRunner.getOrBuildRunner(LazyRunner.java:88) at com.dataiku.dip.eda.compute.engine.LazyRunner.compute(LazyRunner.java:38) at com.dataiku.dip.eda.compute.engine.SmoothlyClosableRunner.compute(SmoothlyClosableRunner.java:23) at com.dataiku.dip.eda.compute.caching.CachedRunner.compute(CachedRunner.java:83) at com.dataiku.dip.eda.worksheets.CardComputeService.computeResultsWithRunner(CardComputeService.java:78) at com.dataiku.dip.eda.worksheets.CardComputeService.computeResults_NT(CardComputeService.java:67) at com.dataiku.dip.eda.worksheets.CardComputeService$1.compute(CardComputeService.java:50) at com.dataiku.dip.eda.worksheets.CardComputeService$1.compute(CardComputeService.java:35) at com.dataiku.dip.futures.SimpleFutureThread.execute(SimpleFutureThread.java:36) at com.dataiku.dip.futures.FutureThreadBase.run(FutureThreadBase.java:88) Caused by: com.dataiku.dip.exceptions.CodedException: EDA local kernel failed to start at com.dataiku.dip.eda.compute.engine.python.LocalPythonKernel.start(LocalPythonKernel.java:33) at com.dataiku.dip.eda.compute.ComputeService.createSampleKernel(ComputeService.java:146) at com.dataiku.dip.eda.compute.ComputeService.access$000(ComputeService.java:52) at com.dataiku.dip.eda.compute.ComputeService$1.call(ComputeService.java:93) at com.dataiku.dip.eda.compute.ComputeService$1.call(ComputeService.java:87) at java.util.concurrent.FutureTask.run(FutureTask.java:266) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) at java.lang.Thread.run(Thread.java:748) Caused by: com.dataiku.dip.io.SocketBlockLink$SecretKernelTimeoutException: Subprocess failed to connect, it probably crashed at startup. Check the logs. at com.dataiku.dip.io.SocketBlockLink.waitForConnection(SocketBlockLink.java:70) at com.dataiku.dip.io.SecretProtectedKernelLink.waitForProcess(SecretProtectedKernelLink.java:39) at com.dataiku.dip.io.SecretProtectedKernelLink.waitForProcess(SecretProtectedKernelLink.java:48) at com.dataiku.dip.io.PythonSecretProtectedKernel.start(PythonSecretProtectedKernel.java:108) at com.dataiku.dip.eda.compute.engine.python.LocalPythonKernel.start(LocalPythonKernel.java:29) ... 8 more Caused by: java.net.SocketException: Socket closed at java.net.PlainSocketImpl.socketAccept(Native Method) at java.net.AbstractPlainSocketImpl.accept(AbstractPlainSocketImpl.java:409) at java.net.ServerSocket.implAccept(ServerSocket.java:545) at java.net.ServerSocket.accept(ServerSocket.java:513) at com.dataiku.dip.io.SocketBlockLink.waitForConnection(SocketBlockLink.java:66) ... 12 more [2020/11/25-19:17:01.509] [qtp1528769018-365630] [INFO] [dip.controllers] - Err for /dip/api/futures/get-update
Best Answer
-
I finally install the missing package directly from the command line, everything is going fine now! The command I run is
dss_data/bin/pip install sortedcontainers
Answers
-
Attached backend logs seem to show that the sortedcontainer package is missing. You can try rebuilding your python built-in environment.
-
I ran the rebuilt command and added this librairy, restart my DSS Instance but still the same error.
Attached a screenshot our installed packages in the python_env
-
The same error as in the backend logs (which was the one I was addressing, about a *local* EDA kernel) or the same error as in you original post's main text (about a *docker* EDA kernel)?
If it's the latter, try unsetting docker configuration and using a local kernel to check whether this works. -
Even with the "local" Kernel, I get the same error
Unexpected EDA kernel failure
com.dataiku.dip.exceptions.CodedException: EDA local kernel failed to start, caused by: CodedException: EDA local kernel failed to start, caused by: SecretKernelTimeoutException: Subprocess failed to connect, it probably crashed at startup. Check the logs., caused by: SocketException: Socket closed
-
Yes, this is the same overall error, the details are in the backend logs you attached the first time, please attache those again.
-
It's the same error I think, as you can see in the attached log file
-
Strange. Can you try, in a local notebook in the built-in environment, to `import sortedcontainer`?
-
Seems that I add it in the python_env, not in the built in Environment!
Because is the Jupyter Notebook, by default, it doesn't import it. But when I change the kernel to my python_environment, all is OK.
What I want is to add this package 'sortedcontainers' to my built in environement!
So how can I add it to the built in Environment so?
-
I am not sure I understand what you mean. I think the easiest would be to share an instance diagnostic with support@dataiku.com (from Administration > Maintenance > Diagnostic tool). If the resulting file is too large for mail (> 15 MB), you can use https://dl.dataiku.com to send it. If so, please don't forget to send the link that is generated when you upload the file.
-
-
You should send that to support@dataiku.com (or on support.dataiku.com), as these files can contain things you may not want public, and support is better armed than community to answer on the specific configuration that you have and that is not common to other users.