Hi, This is just an extra unnecessary usage of the /sparkConf/ member val directly (those two lines are added by two different PRs).
Actually both uses the same /sparkConf/ to give back the config value, as /KubernetesExecutorConf/ extends the /KubernetesConf/ [1] which uses the passed /sparkConf/ to get back the value in the get method [2]. So technically this does not cause any problem but it is better to harmonize it and call the contains method directly on /kubernetesConf/ (see the method [3]) to avoid confusion for the next readers. [1] https://github.com/apache/spark/blob/9b875ceada60732899053fbd90728b4944d1c03d/resource-managers/kubernetes/core/src/main/scala/org/apache/spark/deploy/k8s/KubernetesConf.scala#L132-L138 [2] https://github.com/apache/spark/blob/9b875ceada60732899053fbd90728b4944d1c03d/resource-managers/kubernetes/core/src/main/scala/org/apache/spark/deploy/k8s/KubernetesConf.scala#L67 [3] https://github.com/apache/spark/blob/9b875ceada60732899053fbd90728b4944d1c03d/resource-managers/kubernetes/core/src/main/scala/org/apache/spark/deploy/k8s/KubernetesConf.scala#L65 Best Regards, Attila -- Sent from: http://apache-spark-developers-list.1001551.n3.nabble.com/ --------------------------------------------------------------------- To unsubscribe e-mail: dev-unsubscr...@spark.apache.org
