强制删除spark pod驱动程序时，驱动程序不会重新启动

Question

我有一份spark流媒体工作，我正试图由spark-k8运营商提交。我始终保持重启策略。但是，手动删除驱动程序时，驱动程序不会重新启动。我的yaml：

apiVersion: "sparkoperator.k8s.io/v1beta2"
kind: SparkApplication
metadata:
  name: test-v2
  namespace: default
spec:
  type: Scala
  mode: cluster
  image: "com/test:v1.0"
  imagePullPolicy: Never
  mainClass: com.test.TestStreamingJob
  mainApplicationFile: "local:///opt/spark-2.4.5/work-dir/target/scala-2.12/test-assembly-0.1.jar"
  sparkVersion: "2.4.5"
  restartPolicy:
    type: Always
  volumes:
    - name: "test-volume"
      hostPath:
        path: "/tmp"
        type: Directory
  driver:
    cores: 1
    coreLimit: "1200m"
    memory: "512m"
    labels:
      version: 2.4.5
    serviceAccount: spark
    volumeMounts:
      - name: "test-volume"
        mountPath: "/tmp"
    terminationGracePeriodSeconds: 60
  executor:
    cores: 1
    instances: 2
    memory: "512m"
    labels:
      version: 2.4.5
    volumeMounts:
      - name: "test-volume"
        mountPath: "/tmp"

Spark版本：2.4.5 apiVersion：“sparkoperator.k8s.io/v1beta2”

我遵循的步骤：

通过kubectl apply-f示例/spark测试创建资源。亚马尔。Pod创建成功。手动删除驱动程序。

预期行为：将根据重新启动策略重新启动新的驱动程序吊舱。

实际行为：已删除驱动程序和执行程序吊舱。

环境：在Mac上用Docker测试。具有4个CPU和8 GB内存

来自spark的日志-运算符{驱动程序pod失败，退出代码143，原因：错误}

Answer 1

spark-K8驱动程序有一个问题，现在已经修复，我可以看到手动删除的驱动程序正在重新启动。基本上代码没有处理默认值

https://github.com/GoogleCloudPlatform/spark-on-k8s-operator/pull/898

或者只需配置以下配置，即可不需要默认值“

restartPolicy:
    type: Always
    onFailureRetries: 3
    onFailureRetryInterval: 10
    onSubmissionFailureRetries: 3
    onSubmissionFailureRetryInterval: 10