How to solve a java null pointer exception?
There are times when a process throws a very vague unclear error message and the logs aren't helpful either. For ex:
"The setup does not seem to contain any obvious errors, but you should check the log messages or activate the debug mode in settings dialog in order to get more information about this problem."
First of all, where is the settings dialog? I cannot find a debug mode in the settings menu atleast. Next, the log simply points to the operator where the error occurred and says SEVERE: java.lang.NullPointerException
This could be because 'n' number of things. What would be a general guideline to diagnose these kind of errors?
Thank you.
From searching the forums it seems this error can be thrown for a number of reasons. What is a generic set of practices to go about resolving this issue?
Find more posts tagged with
Thank you Thomas, this definitely helps. Not that it solved my issue but I have a starting point. It would be very helpful in terms of turnaround time if RapidMiner provides more details about such errors in future versions.
No problem, this is one of those old legacy errors that seems to show up every so often. We're looking into it.
I faced this error just recently.
Problem
-----------
The underlying problem with my process was with data types. It was throwing this error while saving the process or storing the final output.
Description
---------------
My process was reading a column that was supposed to have character values. But it wasn't able to read few values. One more symptom of this error was that - it was not able to load/compute stats.
Cause Detection
----------------------
I found the root of the problem by looking at the log file created in the .rapidminer path. It mentioned the column names in the log file. I removed those columns from the process and reran. It was success. Computed stats as well.
I'm still looking into treating those columns.
Altair EmployeeHi,
can you maybe post the stack trace of your log file? That enables us to check the underlying issue
Altair Employeetagging @Marco_Boeck
Hi there, Apologies, I could not reply sooner.
Stack trace file could be difficult to provide now, it was my colleague's machine, hence not reachable. But I did manage to find the issue. Apparently the column had words from greek,chinese and arabic language which was causing this problem. I believe, It was not able to recognize the column type and could not load stats.
Hope this helps.
Regards,
Muddasir
Ah yes, the "obvious errors" message. I usually get this when I want to do clustering and there is something wrong with how I transformed the data. We should probably provide better guidance on that error message.
For now, I would check that meet all the data type and formats pre-conditions for the algorithm you are using. You can right click on each Operator and click "Show Operator Info." In there it should tell you if the algo can do binominal, numerical, etc.
Ah yes, the "obvious errors" message. I usually get this when I want to do clustering and there is something wrong with how I transformed the data. We should probably provide better guidance on that error message.
For now, I would check that meet all the data type and formats pre-conditions for the algorithm you are using. You can right click on each Operator and click "Show Operator Info." In there it should tell you if the algo can do binominal, numerical, etc.