Purushotham Pushpavanthar created SPARK-61267:
-------------------------------------------------
Summary: KubernetesExecutorBackend does not apply ResourceProfile
memory and cores confs to the executor SparkConf
Key: SPARK-61267
URL: https://issues.apache.org/jira/browse/SPARK-61267
Project: Spark
Issue Type: Bug
Components: Kubernetes
Affects Versions: 5.0.0
Reporter: Purushotham Pushpavanthar
SPARK-50421 made CoarseGrainedExecutorBackend.run set spark.executor.memory,
spark.executor.memoryOverhead, spark.memory.offHeap.size and
spark.executor.cores from a non-default ResourceProfile so executor-side
readers (MemoryManager off-heap sizing, PythonRunner worker memory) see the
profile's values. KubernetesExecutorBackend.run duplicates that startup
sequence but was not updated, so Kubernetes executors under a non-default
ResourceProfile still read the application-level values.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
