Github user dvogelbacher commented on a diff in the pull request:
https://github.com/apache/spark/pull/21366#discussion_r192200162
--- Diff:
resource-managers/kubernetes/core/src/main/scala/org/apache/spark/scheduler/cluster/k8s/ExecutorPodsLifecycleEventHandler.scala
---
@@ -0,0 +1,130 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.spark.scheduler.cluster.k8s
+
+import com.google.common.cache.{Cache, CacheBuilder}
+import io.fabric8.kubernetes.api.model.Pod
+import io.fabric8.kubernetes.client.KubernetesClient
+import scala.collection.JavaConverters._
+
+import org.apache.spark.SparkConf
+import org.apache.spark.deploy.k8s.Config._
+import org.apache.spark.deploy.k8s.Constants._
+import org.apache.spark.scheduler.ExecutorExited
+import org.apache.spark.util.Utils
+
+private[spark] class ExecutorPodsLifecycleEventHandler(
+ conf: SparkConf,
+ executorBuilder: KubernetesExecutorBuilder,
+ kubernetesClient: KubernetesClient,
+ podsEventQueue: ExecutorPodsEventQueue,
+ // Use a best-effort to track which executors have been removed
already. It's not generally
+ // job-breaking if we remove executors more than once but it's ideal
if we make an attempt
+ // to avoid doing so. Expire cache entries so that this data structure
doesn't grow beyond
+ // bounds.
+ removedExecutorsCache: Cache[java.lang.Long, java.lang.Long]) {
+
+ import ExecutorPodsLifecycleEventHandler._
+
+ private val eventProcessingInterval =
conf.get(KUBERNETES_EXECUTOR_EVENT_PROCESSING_INTERVAL)
+
+ def start(schedulerBackend: KubernetesClusterSchedulerBackend): Unit = {
+ podsEventQueue.addSubscriber(eventProcessingInterval) { updatedPods =>
+ updatedPods.foreach { updatedPod =>
+ processUpdatedPod(schedulerBackend, updatedPod)
+ }
+ }
+ }
+
+ private def processUpdatedPod(
+ schedulerBackend: KubernetesClusterSchedulerBackend, updatedPod:
Pod) = {
+ val execId =
updatedPod.getMetadata.getLabels.get(SPARK_EXECUTOR_ID_LABEL).toLong
+ if (isDeleted(updatedPod)) {
+ removeExecutorFromSpark(schedulerBackend, updatedPod, execId)
+ } else {
+ updatedPod.getStatus.getPhase.toLowerCase match {
+ // TODO (SPARK-24135) - handle more classes of errors
+ case "error" | "failed" | "succeeded" =>
--- End diff --
btw, I just tried testing a pod with an init error locally and doing
`kubectl get -o=yaml pod {name} | grep phase` gives me `phase: Failed`. So I
think the Init errors of `SPARK-24135` will be handled by this as well.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
