查看namenode地址_我对HDFS的一些理解——HA模式下NameNode启动过程

我对HDFS的一些理解——NameNode启动过程。

public static void main(String argv[]) throws Exception {
  if (DFSUtil.parseHelpArgument(argv, NameNode.USAGE, System.out, true)) {
    System.exit(0);
  }

  try {
    //输出namenode启动的一些信息
    StringUtils.startupShutdownMessage(NameNode.class, argv, LOG);
    //创建一个namenode节点
    NameNode namenode = createNameNode(argv, null);
    if (namenode != null) {
      namenode.join();
    }
  } catch (Throwable e) {
    LOG.error("Failed to start namenode.", e);
    terminate(1, e);
  }
}

上述main函数为namenode的启动入口，StringUtils.startupShutdownMessage主要输出启动该节点的一些系统信息，如classpath，jdk version等等，如下图所示。

接下来主要查看createNameNode(argv, null)做了哪些初始化操作。

public static NameNode createNameNode(String argv[], Configuration conf)
    throws IOException {
  LOG.info("createNameNode " + Arrays.asList(argv));
  //创建conf配置类
  if (conf == null)
    conf = new HdfsConfiguration();
  // Parse out some generic args into Configuration.
  //加载配置
  GenericOptionsParser hParser = new GenericOptionsParser(conf, argv);
  argv = hParser.getRemainingArgs();
  // Parse the rest, NN specific args.
  //获取启动选项
  StartupOption startOpt = parseArguments(argv);
  if (startOpt == null) {
    printUsage(System.err);
    return null;
  }
  setStartupOption(conf, startOpt);

  switch (startOpt) {
    case FORMAT: {
      boolean aborted = format(conf, startOpt.getForceFormat(),
          startOpt.getInteractiveFormat());
      terminate(aborted ? 1 : 0);
      return null; // avoid javac warning
    }
    ...
    default: {
      DefaultMetricsSystem.initialize("NameNode");
      return new NameNode(conf);
    }
  }
}

以上主要是进行HDFS配置文件类HdfsConfiguration创建并加载hdfs-site.xml配置文件中定义的配置。并获取启动选项，根据不同的启动选项在switch分支中进行后续操作，这里是default，即后续代码是newNameNode(conf)。

/**
 * Start NameNode.
 * <p>
 * The name-node can be started with one of the following startup options:
 * <ul> 
 * <li>{@link StartupOption#REGULAR REGULAR} - normal name node startup</li>
 * <li>{@link StartupOption#FORMAT FORMAT} - format name node</li>
 * <li>{@link StartupOption#BACKUP BACKUP} - start backup node</li>
 * <li>{@link StartupOption#CHECKPOINT CHECKPOINT} - start checkpoint node</li>
 * <li>{@link StartupOption#UPGRADE UPGRADE} - start the cluster  
 * <li>{@link StartupOption#UPGRADEONLY UPGRADEONLY} - upgrade the cluster  
 * upgrade and create a snapshot of the current file system state</li> 
 * <li>{@link StartupOption#RECOVER RECOVERY} - recover name node
 * metadata</li>
 * <li>{@link StartupOption#ROLLBACK ROLLBACK} - roll the  
 *            cluster back to the previous state</li>
 * <li>{@link StartupOption#FINALIZE FINALIZE} - finalize 
 *            previous upgrade</li>
 * <li>{@link StartupOption#IMPORT IMPORT} - import checkpoint</li>
 * </ul>
 * The option is passed via configuration field: 
 * <tt>dfs.namenode.startup</tt>
 * 
 * The conf will be modified to reflect the actual ports on which 
 * the NameNode is up and running if the user passes the port as
 * <code>zero</code> in the conf.
 * 
 * @param conf  confirguration
 * @throws IOException
 */
public NameNode(Configuration conf) throws IOException {
  this(conf, NamenodeRole.NAMENODE);
}

主要是this(conf, NamenodeRole.NAMENODE)，说明创建了namenode节点。这里也可以创建其他节点，如namenode的backup node。

rotected NameNode(Configuration conf, NamenodeRole role) 
      throws IOException { 
    this.conf = conf;
    this.role = role;
    //设置可供客户端访问的地址和端口信息。在hdfs-site.xml中配置。日志打印相关设置。
    //NameNode.clientNamenodeAddress属性赋值为hdfsServer
    //<property>
    //    <name>fs.default.name</name>
    //    <value>hdfs://hdfsServer</value>
    //</property>
    setClientNamenodeAddress(conf);
    //nsId值为hdfsServer
    String nsId = getNameServiceId(conf);
    //namenodeId值为nn1
    //<property>
    //    <name>dfs.ha.namenodes.hdfsServer</name>
    //    <value>nn1,nn2</value>
    //</property>
    String namenodeId = HAUtil.getNameNodeId(conf, nsId);
    //true
    this.haEnabled = HAUtil.isHAEnabled(conf, nsId);
    state = createHAState(getStartupOption(conf));
    this.allowStaleStandbyReads = HAUtil.shouldAllowStandbyReads(conf);
    //创建高可用上下文
    this.haContext = createHAContext();
    try {
      initializeGenericKeys(conf, nsId, namenodeId);
      initialize(conf);
      try {
        haContext.writeLock();
        state.prepareToEnterState(haContext);
        state.enterState(haContext);
      } finally {
        haContext.writeUnlock();
      }
    } catch (IOException e) {
      this.stop();
      throw e;
    } catch (HadoopIllegalArgumentException e) {
      this.stop();
      throw e;
    }
    this.started.set(true);
  }

1、首先是setClientNamenodeAddress(conf)，设置了server0:9000相关的配置。如下图所示。

protected void initialize(Configuration conf) throws IOException {
  if (conf.get(HADOOP_USER_GROUP_METRICS_PERCENTILES_INTERVALS) == null) {
    String intervals = conf.get(DFS_METRICS_PERCENTILES_INTERVALS_KEY);
    if (intervals != null) {
      conf.set(HADOOP_USER_GROUP_METRICS_PERCENTILES_INTERVALS,
        intervals);
    }
  }

  UserGroupInformation.setConfiguration(conf);
  loginAsNameNodeUser(conf);

  //指标监控
  NameNode.initMetrics(conf, this.getRole());
  StartupProgressMetrics.register(startupProgress);

  if (NamenodeRole.NAMENODE == role) {
    startHttpServer(conf);
  }

  this.spanReceiverHost =
    SpanReceiverHost.get(conf, DFSConfigKeys.DFS_SERVER_HTRACE_PREFIX);
  
  //加载文件系统
  loadNamesystem(conf);

  rpcServer = createRpcServer(conf);
  if (clientNamenodeAddress == null) {
    // This is expected for MiniDFSCluster. Set it now using 
    // the RPC server's bind address.
    clientNamenodeAddress = 
        NetUtils.getHostPortString(rpcServer.getRpcAddress());
    LOG.info("Clients are to use " + clientNamenodeAddress + " to access"
        + " this namenode/service.");
  }
  if (NamenodeRole.NAMENODE == role) {
    httpServer.setNameNodeAddress(getNameNodeAddress());
    httpServer.setFSImage(getFSImage());
  }
  
  pauseMonitor = new JvmPauseMonitor(conf);
  pauseMonitor.start();
  metrics.getJvmMetrics().setPauseMonitor(pauseMonitor);
  
  startCommonServices(conf);
}