Execution failed while using Create Local Big Data Environment node

I am trying to run association rule for next restaurant prediction from example workflow but I am getting this error

Execute failed: java.lang.RuntimeException: The root scratch dir: C:/Users/372683/AppData/Local/Temp/knime_localspark_6128683598208889476/hive_scratch on HDFS should be writable. Current permissions are: rw-rw-rw-

Is there some setting that I am missing here. Noob here. Pls advise!!

Hi,
I wonder why the permissions are set like that? Spark thrift server creates the folder on-the-fly (via HDFS API) and even sets the expected permissions.
Since this is on Windows, maybe the permissions are inherited from higher up the folder hierarchy? Possibly one of the following folders forces all permission of its subfolders to be rw-rw-rw (this is possible on Windows):
C:/Users/372683/AppData/Local/Temp
C:/Users/372683/AppData/Local/
C:/Users/372683/AppData/
C:/Users/372683/
C:/Users/
Could you try to exempt on of those from permission inheritance, probably the one deepest in the hierarchy should help already: C:/Users/372683/AppData/Local/Temp
Bye
Tobias

1 Like

Hi Tobias,

Windows permissions are weird!!

I was able to run workflow after I changed rights of C:/Users/372683/ folder.

Cheers
Shashank

1 Like

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.