LucaCanali commented on pull request #29477:
URL: https://github.com/apache/spark/pull/29477#issuecomment-702604124


   @dongjoon-hyun I see that this is a minor bug. I'd say that if there is no 
time or necessity to fix this during the current phase of Spark on K8s GA 
preparation, another option could be just to add a note in the documentation, 
with somehting like  "Executors on K8S will not allocate extra memory for 
off-heap allocations based on the configuration `spark.memory.offHeap.size`, 
differently from what happens with executors on YARN. Users will have to 
allocate the extra memory needed for off-heap allocations on K8S using the 
configuration parameter `spark.executorMemoryOverhead` and/or 
`spark.kubernetes.memoryOverheadFactor`."


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to