From 6a4206ff04746481d7c8e307dfd0d31ff1402555 Mon Sep 17 00:00:00 2001 From: Yinan Li Date: Tue, 9 Jan 2018 01:32:48 -0800 Subject: [PATCH] [SPARK-22998][K8S] Set missing value for SPARK_MOUNTED_CLASSPATH in the executors ## What changes were proposed in this pull request? The environment variable `SPARK_MOUNTED_CLASSPATH` is referenced in the executor's Dockerfile, where its value is added to the classpath of the executor. However, the scheduler backend code missed setting it when creating the executor pods. This PR fixes it. ## How was this patch tested? Unit tested. vanzin Can you help take a look? Thanks! foxish Author: Yinan Li Closes #20193 from liyinan926/master. --- .../spark/scheduler/cluster/k8s/ExecutorPodFactory.scala | 5 ++++- .../scheduler/cluster/k8s/ExecutorPodFactorySuite.scala | 3 ++- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/resource-managers/kubernetes/core/src/main/scala/org/apache/spark/scheduler/cluster/k8s/ExecutorPodFactory.scala b/resource-managers/kubernetes/core/src/main/scala/org/apache/spark/scheduler/cluster/k8s/ExecutorPodFactory.scala index 066d7e9f70..bcacb3934d 100644 --- a/resource-managers/kubernetes/core/src/main/scala/org/apache/spark/scheduler/cluster/k8s/ExecutorPodFactory.scala +++ b/resource-managers/kubernetes/core/src/main/scala/org/apache/spark/scheduler/cluster/k8s/ExecutorPodFactory.scala @@ -94,6 +94,8 @@ private[spark] class ExecutorPodFactory( private val executorCores = sparkConf.getDouble("spark.executor.cores", 1) private val executorLimitCores = sparkConf.get(KUBERNETES_EXECUTOR_LIMIT_CORES) + private val executorJarsDownloadDir = sparkConf.get(JARS_DOWNLOAD_LOCATION) + /** * Configure and construct an executor pod with the given parameters. */ @@ -145,7 +147,8 @@ private[spark] class ExecutorPodFactory( (ENV_EXECUTOR_CORES, math.ceil(executorCores).toInt.toString), (ENV_EXECUTOR_MEMORY, executorMemoryString), (ENV_APPLICATION_ID, applicationId), - (ENV_EXECUTOR_ID, executorId)) ++ executorEnvs) + (ENV_EXECUTOR_ID, executorId), + (ENV_MOUNTED_CLASSPATH, s"$executorJarsDownloadDir/*")) ++ executorEnvs) .map(env => new EnvVarBuilder() .withName(env._1) .withValue(env._2) diff --git a/resource-managers/kubernetes/core/src/test/scala/org/apache/spark/scheduler/cluster/k8s/ExecutorPodFactorySuite.scala b/resource-managers/kubernetes/core/src/test/scala/org/apache/spark/scheduler/cluster/k8s/ExecutorPodFactorySuite.scala index 884da8aabd..7cfbe54c95 100644 --- a/resource-managers/kubernetes/core/src/test/scala/org/apache/spark/scheduler/cluster/k8s/ExecutorPodFactorySuite.scala +++ b/resource-managers/kubernetes/core/src/test/scala/org/apache/spark/scheduler/cluster/k8s/ExecutorPodFactorySuite.scala @@ -197,7 +197,8 @@ class ExecutorPodFactorySuite extends SparkFunSuite with BeforeAndAfter with Bef ENV_EXECUTOR_CORES -> "1", ENV_EXECUTOR_MEMORY -> "1g", ENV_APPLICATION_ID -> "dummy", - ENV_EXECUTOR_POD_IP -> null) ++ additionalEnvVars + ENV_EXECUTOR_POD_IP -> null, + ENV_MOUNTED_CLASSPATH -> "/var/spark-data/spark-jars/*") ++ additionalEnvVars assert(executor.getSpec.getContainers.size() === 1) assert(executor.getSpec.getContainers.get(0).getEnv.size() === defaultEnvs.size)