Yarn 之 nodemanager.containermanager.container

src/main/java/org/apache/hadoop/yarn/server/nodemanager/containermanager/container/ContainerState.java

cpp 复制代码
/**
 * States used by the container state machine.
 */
public enum ContainerState {
  // NOTE: In case of future additions / deletions / modifications to this
  //       enum, please ensure that the following are also correspondingly
  //       updated:
  //       1. ContainerImpl::getContainerSubState().
  //       2. the doc in the ContainerSubState class.
  //       3. the doc in the yarn_protos.proto file.
  NEW, LOCALIZING, LOCALIZATION_FAILED, SCHEDULED, RUNNING, RELAUNCHING,
  REINITIALIZING, REINITIALIZING_AWAITING_KILL,
  EXITED_WITH_SUCCESS, EXITED_WITH_FAILURE, KILLING,
  CONTAINER_CLEANEDUP_AFTER_KILL, CONTAINER_RESOURCES_CLEANINGUP, DONE,
  PAUSING, PAUSED, RESUMING
}
cpp 复制代码
      // If this is a recovered container that has already launched, skip
      // uploading resources to the shared cache. We do this to avoid uploading
      // the same resources multiple times. The tradeoff is that in the case of
      // a recovered container, there is a chance that resources don't get
      // uploaded into the shared cache. This is OK because resources are not
      // acknowledged by the SCM until they have been uploaded by the node
      // manager.
      if (container.recoveredStatus != RecoveredContainerStatus.LAUNCHED
          && container.recoveredStatus != RecoveredContainerStatus.COMPLETED) {
        // kick off uploads to the shared cache
        container.dispatcher.getEventHandler().handle(
            new SharedCacheUploadEvent(
                container.resourceSet.getResourcesToBeUploaded(), container
                .getLaunchContext(), container.getUser(),
                SharedCacheUploadEventType.UPLOAD));
      }

      return ContainerState.SCHEDULED;
ContainerLaunch

src/main/java/org/apache/hadoop/yarn/server/nodemanager/containermanager/launcher/ContainerLaunch.java

cpp 复制代码
  protected void handleContainerExitCode(int exitCode, Path containerLogDir) {
    ContainerId containerId = container.getContainerId();
    LOG.debug("Container {} completed with exit code {}", containerId,
        exitCode);

    StringBuilder diagnosticInfo =
        new StringBuilder("Container exited with a non-zero exit code ");
    diagnosticInfo.append(exitCode);
    diagnosticInfo.append(". ");
    if (exitCode == ExitCode.FORCE_KILLED.getExitCode()
        || exitCode == ExitCode.TERMINATED.getExitCode()) {
      // If the process was killed, Send container_cleanedup_after_kill and
      // just break out of this method.
      dispatcher.getEventHandler().handle(
          new ContainerExitEvent(containerId,
              ContainerEventType.CONTAINER_KILLED_ON_REQUEST, exitCode,
              diagnosticInfo.toString()));
    } else if (exitCode != 0) {
      handleContainerExitWithFailure(containerId, exitCode, containerLogDir,
          diagnosticInfo);
    } else {
      LOG.info("Container " + containerId + " succeeded ");
      dispatcher.getEventHandler().handle(
          new ContainerEvent(containerId,
              ContainerEventType.CONTAINER_EXITED_WITH_SUCCESS));
    }
  }
相关推荐
曾阿伦2 天前
Windows 下运行 Hadoop 并部署到 AWS EMR 指南
hadoop·windows·aws
极光代码工作室3 天前
基于Spark的日志监控与分析平台
大数据·hadoop·python·spark·数据可视化
liuxiaowei34 天前
Winform+WPF双框架实战:喷涂工艺SCADA上位机从0到1搭建(附采集监控源码+车间踩坑实录)
大数据·hadoop·wpf
hkNaruto7 天前
【大数据】《传统Hadoop大数据技术入门:补充与进阶——从数据格式到智能决策的问答实录》
大数据·hadoop·分布式
牛奶咖啡138 天前
大数据Hadoop运维应用实践——双NameNode高可用Hadoop集群架构(下)
大数据·hadoop·hadoop集群配置文件解析·hadoop高可用集群安装部署·配置指定多台服务器相互免密·配置hadoop服务开机自启·ansible部署hadoop
李昊哲小课9 天前
spark4 集群安装
大数据·hadoop·zookeeper·spark
buligbulig9 天前
Hadoop环境安装和集群创建
大数据·hadoop·分布式
香山上的麻雀100810 天前
Hive内部表(MANAGED_TABLE)的“批量删除分区”特性详解
数据仓库·hive·hadoop
牛奶咖啡1311 天前
大数据Hadoop运维应用实践——双NameNode高可用Hadoop集群架构(上)
hadoop·高可用hadoop集群搭建·给服务器做免密信任·修改服务器主机名称·服务器本地配置主机名与ip映射·服务器配置jdk环境·服务器安装zooker
jjjava2.012 天前
SpringMVC入门指南:从零掌握核心要点
数据仓库·hive·hadoop