在Hbase中运行命令时，“节点/hbase不在ZooKeeper中”

Question

我正试图以独立模式安装HBase v1.1.2 (使用Hadoopv2.7.2)，并尽可能密切地跟踪Apache HBase文档，但是配置似乎有点棘手。不确定这是否与我用自制软件安装HBase而不是通过他们的镜像站点直接下载的事实有关。我的目标是达到这样的程度，我可以做的事情，如创建表，列出现有表等。

我安装了Java1.7.0，并运行了"brew安装hbase“来安装hbase。在Homebrew的hbase目录中，我打开了/libexec/conf/hbase-env.sh，并在第27行中看到了以下内容：

export JAVA_HOME="$(/usr/libexec/java_home)"

当我在命令行中执行/usr/libexec/java_home时，它输出/Library/Java/JavaVirtualMachines/jdk1.7.0_60.jdk/Contents/Home。因此，似乎有一个有效的$JAVA_HOME env变量正在设置。

在我的/libexec/conf/hbase-site.xml配置文件中，我设置了以下键/值对：

<configuration>
  <property>
    <name>hbase.rootdir</name>
    <value>file:///home/richiethomas/hbase</value>
  </property>
  <property>
    <name>hbase.zookeeper.property.dataDir</name>
    <value>/home/richiethomas/zookeeper</value>
  </property>
</configuration>

换句话说，我已经精确复制了HBase文档中显示的配置，只将'testuser‘替换为’richiethoma‘，因为这是我在命令行中使用的用户名。

然后，在Homebrehbase.sh目录中，运行`.bin/start-hbase.sh‘。我看到以下情况：

starting master, logging to /usr/local/Cellar/hbase/1.1.2/libexec/bin/../logs/hbase-richiethomas-master-MacBook-Pro-5.local.out

然后，运行hbase shell并看到以下内容：

SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/usr/local/Cellar/hbase/1.1.2/libexec/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/usr/local/Cellar/hadoop/2.7.2/libexec/share/hadoop/common/lib/slf4j-log4j12-1.7.10.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]
2017-01-03 17:27:52,056 WARN  [main] util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
HBase Shell; enter 'help<RETURN>' for list of supported commands.
Type "exit<RETURN>" to leave the HBase Shell
Version 1.1.2, rcc2b70cf03e3378800661ec5cab11eb43fafe0fc, Wed Aug 26 20:11:27 PDT 2015

>>

这样做之后，在一个新的CLI选项卡I中，我看到了上面看到的路径( cat )中的日志，我看到了以下内容：

SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/usr/local/Cellar/hbase/1.1.2/libexec/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/usr/local/Cellar/hadoop/2.7.2/libexec/share/hadoop/common/lib/slf4j-log4j12-1.7.10.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]

然后，回到我的HBase shell中，运行以下命令以查看是否一切正常，并看到以下内容：

>> list
TABLE                                                                                                                                                                            
2017-01-03 17:28:05,108 ERROR [main] client.ConnectionManager$HConnectionImplementation: The node /hbase is not in ZooKeeper. It should have been written by the master. Check the value configured in 'zookeeper.znode.parent'. There could be a mismatch with the one configured in the master.
2017-01-03 17:28:05,320 ERROR [main] client.ConnectionManager$HConnectionImplementation: The node /hbase is not in ZooKeeper. It should have been written by the master. Check the value configured in 'zookeeper.znode.parent'. There could be a mismatch with the one configured in the master.
2017-01-03 17:28:05,625 ERROR [main] client.ConnectionManager$HConnectionImplementation: The node /hbase is not in ZooKeeper. It should have been written by the master. Check the value configured in 'zookeeper.znode.parent'. There could be a mismatch with the one configured in the master.
2017-01-03 17:28:06,128 ERROR [main] client.ConnectionManager$HConnectionImplementation: The node /hbase is not in ZooKeeper. It should have been written by the master. Check the value configured in 'zookeeper.znode.parent'. There could be a mismatch with the one configured in the master.
2017-01-03 17:28:07,134 ERROR [main] client.ConnectionManager$HConnectionImplementation: The node /hbase is not in ZooKeeper. It should have been written by the master. Check the value configured in 'zookeeper.znode.parent'. There could be a mismatch with the one configured in the master.
2017-01-03 17:28:09,144 ERROR [main] client.ConnectionManager$HConnectionImplementation: The node /hbase is not in ZooKeeper. It should have been written by the master. Check the value configured in 'zookeeper.znode.parent'. There could be a mismatch with the one configured in the master.
2017-01-03 17:28:13,158 ERROR [main] client.ConnectionManager$HConnectionImplementation: The node /hbase is not in ZooKeeper. It should have been written by the master. Check the value configured in 'zookeeper.znode.parent'. There could be a mismatch with the one configured in the master.

ERROR: The node /hbase is not in ZooKeeper. It should have been written by the master. Check the value configured in 'zookeeper.znode.parent'. There could be a mismatch with the one configured in the master.

Here is some help for this command:
List all tables in hbase. Optional regular expression parameter could
be used to filter the output. Examples:

  hbase> list
  hbase> list 'abc.*'
  hbase> list 'ns:abc.*'
  hbase> list 'ns:.*'

根据Apache文档：

您不需要创建HBase数据目录。HBase会为你这么做的。如果创建目录，HBase将尝试进行迁移，这不是您想要的。

我将此解释为不需要在hbase-site.xml中设置以下属性，特别是因为'hbase‘是’zookeeper.znode.parent‘的默认设置：

<property>
    <name>zookeeper.znode.parent</name>
    <value>/hbase</value>
</property>

jps的输出如下：

32146 ZooKeeperMain
1364 QuorumPeerMain
83568 
34120 Jps

我的zoo.cfg文件位于/usr/local/etc/zookeeper/zoo.cfg中，如下所示：

# The number of milliseconds of each tick
tickTime=2000
# The number of ticks that the initial 
# synchronization phase can take
initLimit=10
# The number of ticks that can pass between 
# sending a request and getting an acknowledgement
syncLimit=5
# the directory where the snapshot is stored.
# do not use /tmp for storage, /tmp here is just 
# example sakes.
dataDir=/usr/local/var/run/zookeeper/data
# the port at which the clients will connect
clientPort=2181
# the maximum number of client connections.
# increase this if you need to handle more clients
#maxClientCnxns=60
#
# Be sure to read the maintenance section of the 
# administrator guide before turning on autopurge.
#
# http://zookeeper.apache.org/doc/current/zookeeperAdmin.html#sc_maintenance
#
# The number of snapshots to retain in dataDir
#autopurge.snapRetainCount=3
# Purge task interval in hours
# Set to "0" to disable auto purge feature
#autopurge.purgeInterval=1

我尝试过的解决方案：

I.

我尝试将hbase-site.xml文件中的hbase-site.xml值替换为/usr/local/var/run/zookeeper/data，但这并没有解决问题。

II.

当我在命令行中运行zkcli并运行ls /时，我只看到zookeeper，而没有看到hbase目录。我尝试在zookeeper中运行以下命令：

create /hbase my_data

然后我看到以下情况：

[zk: localhost:2181(CONNECTED) 4] ls /
[hbase, zookeeper]

然后，当我退出并重新运行HBase shell并尝试list时，我会看到以下内容：

>> list
TABLE                                                                                                                                                                            

ERROR: Can't get master address from ZooKeeper; znode data == null

III.

根据这篇Quora文章，HBase 1.1.2只支持2.5.x版本的Hadoop，所以我运行brew uninstall hadoop和brew install hadoop25，运行停止并启动hbase脚本，并重新启动hbase。还是没有运气。

IV.

我尝试卸载Homebrew并从Apache镜像站点下载HBase，并重复hbase-site.xml中的配置过程，但我得到了相同的错误(The node /hbase is not in ZooKeeper. It should have been written by the master.)。

V.

我通过Homebrew卸载/重新安装了HBase，并遵循了“注意事项”部分中的说明：

==> Caveats
To have launchd start hbase now and restart at login:
  brew services start hbase
Or, if you don't want/need a background service you can just run:
  /usr/local/opt/hbase/bin/start-hbase.sh

我注意到，在运行上述命令之后，hbase-site.xml文件中填充了hbase.rootdir和hbase.zookeeper.property.dataDir的看似合理的值(特别是file:///usr/local/var/hbase和/usr/local/var/zookeeper )。我还注意到，当我运行brew services start hbase，然后运行jps时，我没有看到一个' HMaster‘进程在运行，但是当我运行/usr/local/opt/hbase/bin/start-hbase.sh时，我确实看到了HMaster。但是，我随后尝试启动hbase shell并键入status，并得到了与以前相同的错误(The node /hbase is not in ZooKeeper.)。

六.

我已经验证了我的Hadoop安装是否运行正常：

MacBook-Pro-5:2.5.2 richiethomas$ hadoop jar /usr/local/Cellar/hadoop25/2.5.2/libexec/share/hadoop/mapreduce/hadoop-mapreduce-examples-2.5.2.jar pi 2 5
Number of Maps  = 2
Samples per Map = 5
17/01/03 20:09:33 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Wrote input for Map #0
Wrote input for Map #1
Starting Job
...
Job Finished in 2.156 seconds
Estimated value of Pi is 3.60000000000000000000

我的HBase配置似乎有问题，但我不确定这里还可以尝试什么。

Answer 1

由于您希望设置一个独立的hbase，因此不需要构建hdfs或zk。

问题的原因是，hbase将尝试启动内部zk服务，但由于已经存在java进程QuorumPeerMain占用端口2181，hbase将无法启动。如果hbase成功启动，将有一个名为HMaster的java进程。

要验证这一点，请检查日志hbase-richiethomas-master-MacBook-Pro-5.local.out的最后一行，您可能会看到以下错误消息：

Could not start ZK at requested port of 2181.  ZK was started at port: 2182.  Aborting as clients (e.g. shell) will not be able to find this ZK quorum.

因此，解决办法是：

1. 杀死现有的QuorumPeerMain
2. 再次运行start-hbase.sh