Hadoop2.7.1 HA集群部署

零點

2016-12-18

关注关注

1.修改文件/etc/hostname里的值即可，修改成功后用hostname命令查看当前主机名是否设置成功。

[root@masternode CentOS]# cat /etc/hosts

#127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4

#::1 localhost localhost.localdomain localhost6 localhost6.localdomain6

10.10.10.3 slavenode1.novalocal slavenode1

10.10.10.4 masternode.novalocal masternode

10.10.10.5 slavenode2.novalocal slavenode2

10.10.10.6 slavenode3.novalocal slavenode3

修改hostname

vi /etc/sysconfig/network

Hostname=masternode1

3.在主节点机器上设置ssh免密码登陆

1) 首先在主机器上核对ssh是否安装

[root@masternode ~]# rpm -qa |grep ssh

libssh2-1.4.2-1.el6.x86_64

openssh-5.3p1-104.el6_6.1.x86_64

openssh-server-5.3p1-104.el6_6.1.x86_64

openssh-clients-5.3p1-104.el6_6.1.x86_64

2) 生产密钥

[root@masternode ~]# cd .ssh/

[root@masternode .ssh]# ls

authorized_keys

[root@masternode .ssh]# cd /

[root@masternode /]# ssh-keygen -t rsa

Generating public/private rsa key pair.

Enter file in which to save the key (/opt/.ssh/id_rsa):

Enter passphrase (empty for no passphrase):

Enter same passphrase again:

Your identification has been saved in /opt/.ssh/id_rsa.

Your public key has been saved in /opt/.ssh/id_rsa.pub.

The key fingerprint is:

e8:3d:75:11:0b:6a:a9:f5:39:e5:04:71:2e:94:21:94 [email protected]

The key's randomart image is:

+--[ RSA 2048]----+

| .o.*+o |

| E=.= o |

| = . * |

| = . * . |

| o S = o |

| . . . o |

| . o |

| . |

| |

+-----------------+

[root@masternode /]# cd

[root@masternode ~]# cd .ssh/

[root@masternode .ssh]# ls

authorized_keys id_rsa id_rsa.pub 生产的密钥

[root@masternode .ssh]# cat id_rsa.pub >> authorized_keys

3)把密钥传输到其他节点机器上

（1）用ssh-copy-id命令将公钥传送到远程主机上(这里以Slave1node3为例)。

[root@masternode ~]# ssh-copy-id root@slavenode3

（2）如果在用命令ssh-copy-id时发现找不到该命令“ssh-copy-id：Command not found”，则可能是ssh服务的版本太低的原因，比如若你的机器是RedHat系统就可能该问题，解决办法是：手动复制本地的pubkey内容到远程服务器，命令如下：

cat ~/.ssh/id_rsa.pub | ssh root@slavenode3 'cat >> ~/.ssh/authorized_keys'

该命令等价于下面两个命令：

①在本地机器上执行：scp ~/.ssh/id_rsa.pub root@slavenode3:/~

②到远程机器上执行：cat ~/id_rsa.pub >> ~/.ssh/authorized_keys

[root@masternode .ssh]# scp authorized_keys [email protected]:/opt/.ssh

The authenticity of host '125.208.30.89(125.208.30.89)' can't be established.

RSA key fingerprint is e3:97:c0:29:e4:fa:0d:41:31:6e:df:fe:0c:6b:c7:08.

Are you sure you want to continue connecting (yes/no)? yes

Warning: Permanently added '125.208.30.89' (RSA) to the list of known hosts.

Permission denied (publickey,gssapi-keyex,gssapi-with-mic).

lost connection

[root@masternode .ssh]# vi authorized_keys

cat id_rsa.pub >> authorized_keysd_rsa.pub

[root@masternode .ssh]# scp .ssh/authorized_keys root@mslavenode1:~/.ssh/

[root@masternode .ssh]# scp .ssh/authorized_keys root@slavenode2:~/.ssh/

[root@masternode .ssh]# scp .ssh/authorized_keys root@slavenode3:~/.ssh/

4.把主机hosts拷贝到其他从机器上

[root@masternode centos]# scp /etc/hosts [email protected]:/etc/hosts

hosts 100% 332 0.3KB/s 00:00

[root@masternode centos]# scp /etc/hosts [email protected]:/etc/hosts

hosts 100% 332 0.3KB/s 00:00

[root@masternode centos]# scp /etc/hosts [email protected]:/etc/hosts

5.安装java软件

1)解压java软件包

cd /usr/java/

tar -xvf jdk-7u79-linux-x64.tar.gz

chown -R Hadoop:hadoop jdk-7u79

2）编辑环境变量

编辑"/etc/profile"文件，在后面添加Java的"JAVA_HOME"、"CLASSPATH"以及"PATH"内容如下：

vi /etc/profile

export JAVA_HOME=/usr/java/jdk1.7.0_79

export CLASSPATH=.:$JAVA_HOME/jre/lib/rt.jar:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar

#export JAVA_LIBRARY_PATH='/opt/hadoop/hadoop-2.7.2/lib/native'

export PATH=$PATH:$JAVA_HOME/bin

#set hadoop path

export HADOOP_HOME=/opt/hadoop/hadoop-2.7.2

#export HADOOP_COMMON_LIB_NATIVE_DIR=$HADOOP_HOME/lib/native

#export HADOOP_OPTS="-Djava.library.path=$HADOOP_HOME/lib:$HADOOP_COMMON_LIB_NATIVE_DIR"

export PATH=$PATH:$HADOOP_HOME/bin

#set hive

export HIVE_HOME=/opt/hadoop/hive-2.0.0

export PATH=$PATH:$HIVE_HOME/bin

#set zookeeper

export ZOOKEEPER_HOME=/opt/hadoop/zookeeper

export PATH=$PATH:$ZOOKEEPER_HOME/bin:$ZOOKEEPER_HOME/conf

#set hbase

export HBASE_HOME=/opt/hadoop/hbase-1.1.5

export PATH=$PATH:$HBASE_HOME/bin

#set scala

export SCALA_HOME=/opt/hadoop/scala

export PATH=$PATH:$SCALA_HOME/bin

#set spark

export SPARK_HOME=/opt/hadoop/spark

export PATH=$PATH:$SPARK_HOME/bin

3）使配置生效

保存并退出，执行下面命令使其配置立即生效。

source /etc/profile 或 . /etc/profile

vi ~/.bash_profile

export JAVA_HOME=/usr/java/jdk1.7.0_79/

export CLASSPATH=.:$JAVA_HOME/jre/lib/rt.jar:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar

export PATH=$PATH:$JAVA_HOME/bin

#set hadoop path

export HADOOP_HOME=/opt/hadoop/hadoop-2.7.2

export PATH=$PATH:$HADOOP_HOME/bin

Source ~/.bash_profile

4）验证安装成功

配置完毕并生效后，用下面命令判断是否成功。

[root@masternode java]# java -version

java version "1.7.0_79"

Java(TM) SE Runtime Environment (build 1.7.0_79-b15)

Java HotSpot(TM) 64-Bit Server VM (build 24.79-b02, mixed mode)

5）安装剩余机器java

通过scp命令格式把"/usr/java/"文件复制到其他Slave上面，剩下的事儿就是在其余的Slave服务器上按照上面步骤配置环境变量和测试是否安装成功，这里以Slavenode2为例：

[root@masternode java]# scp -r jdk1.7.0_79/ [email protected]:/usr/java/

此时不可以选择较低版本的JDK进行安装，因为所有集群中的JDK版本必须相同

Hadoop集群之zookeeper安装配置-hbase安装就不用安装这个模块

1．下��地址：

http://mirror.bjtu.edu.cn/apache/zookeeper/zookeeper-3.4.8/zookeeper-3.4.8.tar.gz
2.zookeepe 安装

①　上传软件并解压

把zookeeper-3.4.8.tar.gz文件存放在/opt/hadoop目录下，进行解压：

[root@masternode hadoop]# tar -zxvf zookeeper-3.4.8.tar.gz

mv zookeeper-3.4.8 zookeeper

---要是单独有自己的用户名则执行下面命令

hadoop@Ubuntu:~$ sudo tar -zxvf zookeeper-3.4.8.tar.gz

hadoop@ubuntu:~$ chown -R hadoop:hadoop zookeeper-3.4.8

mv zookeeper-3.4.8 zookeeper

②　设置环境变量 vi /etc/profile

#set zookeeper

export ZOOKEEPER_HOME=/opt/hadoop/zookeeper

export PATH=$PATH:$ZOOKEEPER_HOME/bin

Vi ~/.bash_profile

#set zookeeper

export ZOOKEEPER_HOME=/opt/hadoop/zookeeper

export PATH=$PATH:$ZOOKEEPER_HOME/bin

[root@masternode hadoop]# source /etc/profile

[root@masternode hadoop]# cd zookeeper/conf/

[root@masternode conf]# pwd

/opt/hadoop/zookeeper/conf

③　配置zoo.cfg文件

配置文件存放在$ZOOKEEPER_HOME/conf/目录下，将zoo_sample.cfd文件名称改为zoo.cfg, 缺省的配置内容如下：

[root@masternode conf]# mv zoo_sample.cfg zoo.cfg

vi zoo.cfg

clientPort=2181

dataDir=/opt/hadoop/zookeeper/data/

dataLogDir=/opt/hadoop/zookeeper/log/

server.1=slavenode1:2888:3888

server.2=masternode:2888:3888

server.3=slavenode2:2888:3888

server.4=slavenode3:2888:3888

配置说明：
tickTime：这个时间是作为 Zookeeper 服务器之间或客户端与服务器之间维持心跳的时间间隔，也就是每个 tickTime 时间就会发送一个心跳。
dataDir：顾名思义就是 Zookeeper 保存数据的目录，默认情况下，Zookeeper 将写数据的日志文件也保存在这个目录里。
clientPort：这个端口就是客户端连接 Zookeeper 服务器的端口，Zookeeper 会监听这个端口，接受客户端的访问请求。

④　新建两个目录

mkdir /opt/hadoop/zookeeper/data/

mkdir /opt/hadoop/zookeeper/log/

在 /opt/hadoop/zookeeper/data/目录下创建一个文件：myid

touch /opt/hadoop/zookeeper/data/myid

[root@masternode data]# echo ‘1’> myid

⑤　将zookeeper目录和环境变量设置文件拷贝到其他机器上

[root@masternode hadoop]# scp -r zookeeper root@slavenode1:/opt/hadoop/

[root@masternode hadoop]# scp -r zookeeper root@slavenode3:/opt/hadoop/

[root@masternode hadoop]# scp -r zookeeper root@slavenode2:/opt/hadoop/

[root@masternode conf]# scp ~/.bash_profile root@slavenode2:~/.bash_profile

[root@masternode conf]# scp ~/.bash_profile root@slavenode3:~/.bash_profile

[root@masternode conf]# scp ~/.bash_profile root@slavenode1:~/.bash_profile

⑥　修改其他机器该文件

[root@masternode centos]# cat /opt/hadoop/zookeeper/data/myid

[root@slavenode3 centos]# cat /opt/hadoop/zookeeper/data/myid

⑦　启动zookeeper (所配的机器都要启动)

cd /opt/hadoop/zookeeper/bin

./zkServer.sh start

⑧　验证

[root@masternode zookeeper]# jps

16159 HQuorumPeer

14438 DataNode

16397 Jps

14549 NodeManager

查看状态

[root@slavenode3 centos]# /opt/hadoop/zookeeper/bin/zkServer.sh status

ZooKeeper JMX enabled by default

Using config: /opt/hadoop/zookeeper/bin/../conf/zoo.cfg

Mode: leader

客户端链接zookeeper

[root@slavenode3 centos]# /opt/hadoop/zookeeper/bin/zkCli.sh -server slavenode3:2181

WATCHER::

WatchedEvent state:SyncConnected type:None path:null

[zk: slavenode3:2181(CONNECTED) 0] ls /

[hbase, zookeeper]

[zk: slavenode3:2181(CONNECTED) 1] ls /hbase

[meta-region-server, backup-masters, region-in-transition, table, draining, table-lock, running, master, namespace, hbaseid, online-snapshot, replication, splitWAL, recovering-regions, rs, flush-table-proc]

安装hadoop

1.安装hadoop软件包

①　解压hadoop软件包

Cd /opt/hadoop/

tar -xvf hadoop-2.7.2.tar.gz

[root@masternode hadoop]# ls

hadoop-2.7.2 jdk1.7.0_79 lost+found

hadoop-2.7.2.tar.gz jdk-7u79-linux-x64.tar.gz

[root@masternode hadoop]# cd hadoop-2.7.2

[root@masternode hadoop-2.7.2]# ls

bin include libexec NOTICE.txt sbin

etc lib LICENSE.txt README.txt share

②　设置环境变量

并把Hadoop的安装路径添加到"/etc/profile"中，修改"/etc/profile"文件，将以下语句添加到末尾，并使其生效(. /etc/profile)：

#set hadoop path

export HADOOP_HOME=/opt/hadoop/hadoop-2.7.2

export PATH=$PATH:$HADOOP_HOME/bin

[root@masternode conf]# hadoop version

Hadoop 2.7.2

Subversion https://git-wip-us.apache.org/repos/asf/hadoop.git -r b165c4fe8a74265c792ce23f546c64604acf0e41

Compiled by jenkins on 2016-01-26T00:08Z

Compiled with protoc 2.5.0

From source with checksum d0fda26633fa762bff87ec759ebe689c

This command was run using /opt/hadoop/hadoop-2.7.2/share/hadoop/common/hadoop-common-2.7.2.jar

③　创建临时目录

[root@masternode hadoop-2.7.2]# mkdir -p /opt/hadoop/tmp

[root@masternode hadoop-2.7.2]# mkdir -p /opt/hadoop/hdfs/data

[root@masternode hadoop-2.7.2]# mkdir -p /opt/hadoop/hdfs/name

mkdir /opt/hadoop/hadoop-2.7.2/pids

mkdir /opt/hadoop/tmp/journal

chmod 755 /opt/hadoop/hdfs/data

chmod 755 /opt/hadoop/hdfs/name

chmod 755 /opt/hadoop/tmp

chmod 755 /opt/hadoop/tmp/journal

④　配置hadoop

修改配置文件

需要修改的配置文件主要有下面几个：

./core-site.xml

./hdfs-site.xml

./mapred-site.xml（开始的安装包中是没有这个文件的，需要将文

件./mapred-site.xml.template复制并且重名为mapred-site.xml）。

./yarn-site.xml

./yarn-env.sh

./hadoop-env.sh

./slave

[root@masternode hadoop-2.7.2]# pwd

/opt/hadoop/hadoop-2.7.2

[root@masternode hadoop-2.7.2]# ls

bin include libexec NOTICE.txt sbin

etc lib LICENSE.txt README.txt share

[root@masternode hadoop-2.7.2]# cd etc/hadoop/

/opt/hadoop/hadoop-2.7.2/etc/hadoop

（1）配置hadoop-env.sh

配置 hadoop-env.sh文件-->修改JAVA_HOME

export JAVA_HOME=/usr/java/jdk1.7.0_79/

export HADOOP_PID_DIR=$HADOOP_HOME/pids

export HADOOP_HOME=/opt/hadoop/hadoop-2.7.2

export HADOOP_PID_DIR=$HADOOP_HOME/pids

export PATH=$PATH:$HADOOP_HOME/bin

(2）配置yarn-env.sh文件

配置 yarn-env.sh 文件-->>修改JAVA_HOME

# some Java parameters

export JAVA_HOME=/usr/java/jdk1.7.0_79/

export YARN_PID_DIR=/opt/hadoop/hadoop-2.7.2/pids

（3）配置slaves

配置slaves文件-->>增加slave节点

slavenode2

slavenode3

(4)配置 core-site.xml文件

<name>fs.trash.interval</name>

</property>

<name>fs.defaultFS</name>

<value>hdfs://cluster-ha</value>

</property>

<name>dfs.journalnode.edits.dir</name>

<value>/opt/hadoop/tmp/journal</value>

</property>

<name>io.file.buffer.size</name>

</property>

<name>hadoop.tmp.dir</name>

<value>/opt/hadoop/tmp</value>

</property>

<name>ha.zookeeper.quorum</name>

<value>slavenode1:2181,slavenode3:2181,slavenode2:2181</value>

</property>

<name>ha.zookeeper.session-timeout.ms</name>

</property>

</configuration>

(5)配置hdfs-site.xml 文件-->>增加hdfs配置信息（namenode、datanode端口和目录位置）

<name>dfs.nameservices</name>

<value>cluster-ha</value>

</property>

<name>dfs.ha.namenodes.cluster-ha</name>

</property>

<name>dfs.namenode.rpc-address.cluster-ha.nn</name>

<value>slavenode1:9000</value>

</property>

<name>dfs.namenode.http-address.cluster-ha.nn</name>

<value>slavenode1:50070</value>

</property>

<name>dfs.namenode.rpc-address.cluster-ha.snn</name>

<value>masternode:9000</value>

</property>

<name>dfs.namenode.http-address.cluster-ha.snn</name>

<value>masternode:50070</value>

</property>

<name>dfs.namenode.shared.edits.dir</name>

<value>qjournal://slavenode1:8485;slavenode2:8485;slavenode3:8485;masternode:8485/cluster-ha</value>

</property>

<name>dfs.client.failover.proxy.provider.cluster-ha</name>

<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>

</property>

<name>dfs.ha.fencing.methods</name>

<value>sshfence</value>

</property>

<name>dfs.journalnode.edits.dir</name>

<value>/opt/hadoop/tmp/journal</value>

</property>

<name>dfs.ha.automatic-failover.enabled</name>

</property>

<name>dfs.namenode.name.dir</name>

<value>file:/opt/hadoop/hdfs/name</value>

</property>

<name>dfs.datanode.data.dir</name>

<value>file:/opt/hadoop/hdfs/data</value>

</property>

<name>dfs.replication</name>

</property>

<name>dfs.webhdfs.enabled</name>

</property>

<name>ha.zookeeper.quorum</name>

<value>slavenode1:2181,slavenode3:2181,slavenode2:2181</value>

</property>

<name>dfs.namenode.handler.count</name>

<description>The number of server threads for the namenode.</description>

</property>

<name>dfs.datanode.handler.count</name>

<description>The number of server threads for the datanode.</description>

</property>

<name>dfs.client.socket-timeout</name>

</property>

<name>dfs.datanode.max.transfer.threads</name>

</property>

<property>
<name>dfs.ha.fencing.methods</name>
<value>sshfence
shell(/bin/true)</value>
</property>
<property>
<name>dfs.ha.fencing.ssh.private-key-files</name>
<value>/home/hadoop/.ssh/id_rsa</value>
</property>
<property>
<name>dfs.ha.fencing.ssh.connect-timeout</name>
<value>30000</value>
</property>

</configuration>

(6)配置 mapred-site.xml 文件

-->>增加mapreduce配置（使用yarn框架、jobhistory使用地址以及web地址）

[root@masternode hadoop]# mv mapred-site.xml.template mapred-site.xml

vi mapred-site.xml

<name>mapreduce.framework.name</name>

</property>

<name>mapreduce.job.maps</name>

</property>

<name>mapreduce.job.reduces</name>

</property>

</configuration>

(7)配置yarn-site.xml 文件-->>增加yarn功能

<name>yarn.log-aggregation-enable</name>

</property>

<name>yarn.log-aggregation.retain-seconds</name>

</property>

<name>yarn.resourcemanager.zk-address</name>

<value>slavenode1:2181,slavenode3:2181,slavenode2:2181</value>

</property>

<name>yarn.resourcemanager.cluster-id</name>

<value>cluster-yarn</value>

</property>

<name>yarn.resourcemanager.ha.enabled</name>

</property>

<name>yarn.resourcemanager.ha.rm-ids</name>

</property>

<name>yarn.resourcemanager.hostname.rm1</name>

<value>slavenode1</value>

</property>

<name>yarn.resourcemanager.hostname.rm2</name>

<value>masternode</value>

</property>

<name>yarn.resourcemanager.scheduler.address.rm1</name>

<value>slavenode1:8030</value>

</property>

<name>yarn.resourcemanager.scheduler.address.rm2</name>

<value>masternode:8030</value>

</property>

<name>yarn.resourcemanager.resource-tracker.address.rm1</name>

<value>slavenode1:8031</value>

</property>

<name>yarn.resourcemanager.resource-tracker.address.rm2</name>

<value>masternode:8031</value>

</property>

<name>yarn.resourcemanager.address.rm1</name>

<value>slavenode1:8032</value>

</property>

<name>yarn.resourcemanager.address.rm2</name>

<value>masternode:8032</value>

</property>

<name>yarn.resourcemanager.admin.address.rm1</name>

<value>slavenode:8033</value>

</property>

<name>yarn.resourcemanager.admin.address.rm2</name>

<value>masternode:8033</value>

</property>

<name>yarn.resourcemanager.webapp.address.rm1</name>

<value>slavenode1:8088</value>

</property>

<name>yarn.resourcemanager.webapp.address.rm2</name>

<value>masternode:8088</value>

</property>

<name>yarn.resourcemanager.ha.automatic-failover.enabled</name>

</property>

<name>yarn.resourcemanager.ha.automatic-failover.embedded</name>

</property>

<name>yarn.resourcemanager.ha.automatic-failover.zk-base-path</name>

<value>/yarn-leader-election</value>

</property>

<name>yarn.resourcemanager.recovery.enabled</name>

</property>

<name>yarn.resourcemanager.store.class</name>

<value>org.apache.hadoop.yarn.server.resourcemanager.recovery.ZKRMStateStore</value>

</property>

<name>yarn.nodemanager.aux-services</name>

<value>mapreduce_shuffle</value>

</property>

<name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name>

<value>org.apache.hadoop.mapred.ShuffleHandler</value>

</property>

</configuration>

[root@slavenode1 hadoop-2.7.2]# vi etc/hadoop/log4j.properties

log4j.logger.org.apache.hadoop.util.NativeCodeLoader=ERROR

为了解决启动进程会出现报错

现在在Master机器上的Hadoop配置就结束了，剩下的就是配置Slave机器上的Hadoop。

最简单的方法是将 Master上配置好的hadoop所在文件夹"/opt/hadoop/hadoop-2.7.2"复制到所有的Slave的"/opt/hadoop"目录下（实际上Slave机器上的slavers文件是不必要的，复制了也没问题）。用下面命令格式进行。（备注：此时用户可以为普通用户也可以为root）

7.将配置好的hadoop文件copy到另一台slave机器上

[root@masternode hadoop]# scp -r hadoop-2.7.2/ [email protected]:/opt/hadoop/

接着在"Slave1 .Hadoop"上修改"/etc/profile"文件，将以下语句添加到末尾，并使其有效（source /etc/profile）：

export HADOOP_HOME=/opt/hadoop/hadoop-2.7.2

export PATH=$PATH:$HADOOP_HOME/bin

保存并退出，执行下面命令使其配置立即生效��

source /etc/profile 或 . /etc/profile

8.启动及验证

1. 格式化namenode:

[root@masternode hadoop]# pwd

/opt/hadoop

[root@masternode hadoop]# cd hadoop-2.7.2/bin

hdfs zkfc –formatZK--格式化zookeeper custer-ha主目录

2. 启动启动namenode日志同步服务journalnode（四台机器）

/home/hadoop/hadoop-2.7.2/sbin/hadoop-daemon.sh start journalnode

3. 启动namenode、同步备用namenode、启动备用namenode

①　对NameNode（masternode）节点进行格式化

# hadoop namenode -format

注意：格式化第二次有可能会造成DataNode无法启动，原因是NameSpaceID不一致造成，解决方法是找出不一致的VERSION修改NameSpaceID，也可以尝试删除hdfs/data目录。

②　启动masternode（active）节点NameNode

d /opt/hadoop/hadoop-2.7.2/sbin

# hadoop-daemon.sh start namenode

③　Slavenode1节点上同步（masternode）元数据

# hdfs namenode -bootstrapStandby

④　在masternode格式化ZKFC

# hdfs zkfc -formatZK

⑤　在masternode节点启动HDFS集群

# start-dfs.sh

⑥　在Master0节点启动YARN集群

# start-yarn.sh

⑦　在Master1(slavenode1)节点启动RM

# yarn-daemon.sh start resourcemanager

⑧　以后也可以通过start-all.sh与stop-all.sh启停Hadoop集群

[root@masternode hadoop-2.7.2]# ./bin/hdfs namenode -format

4. 启动hadoop

root@masternode centos]# cd /opt/hadoop/hadoop-2.7.2

[root@masternode hadoop-2.7.2]# sbin/start-all.sh

[root@slavenode1 hadoop-2.7.2]# sbin/stop-all.sh

This script is Deprecated. Instead use stop-dfs.sh and stop-yarn.sh

16/09/19 01:40:10 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable 有警告

Stopping namenodes on [slavenode1]

slavenode1: stopping namenode

slavenode3: stopping datanode

slavenode2: stopping datanode

masternode: stopping datanode

Stopping secondary namenodes [slavenode1]

slavenode1: stopping secondarynamenode

16/09/19 01:40:29 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable

stopping yarn daemons

stopping resourcemanager

slavenode2: stopping nodemanager

masternode: stopping nodemanager

slavenode3: stopping nodemanager

no proxyserver to stop

[root@slavenode1 hadoop-2.7.2]# ldd lib/native/libhadoop.so.1.0.0

lib/native/libhadoop.so.1.0.0: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by lib/native/libhadoop.so.1.0.0)

linux-vdso.so.1 => (0x00007fff63bff000)

libdl.so.2 => /lib64/libdl.so.2 (0x00007fc1127c5000)

libc.so.6 => /lib64/libc.so.6 (0x00007fc112430000)

/lib64/ld-linux-x86-64.so.2 (0x00007fc112bef000)

[root@slavenode1 hadoop-2.7.2]# ldd --version

ldd (GNU libc) 2.12

This is free software; see the source for copying conditions. There is NO

warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Written by Roland McGrath and Ulrich Drepper.

原来系统预装的glibc库是2.12版本，而hadoop期望是2.14版本，所以打印警告信息。

现在有两个办法，重新编译glibc.2.14版本，安装后专门给hadoop使用，这个有点危险。

第二个办法直接在log4j日志中去除告警信息。

[root@slavenode1 hadoop-2.7.2]# vi etc/hadoop/log4j.properties

log4j.logger.org.apache.hadoop.util.NativeCodeLoader=ERROR

9.验证hadoop

（1）验证方法一：用"jps"命令

在Master上用 java自带的小工具jps查看进程。

[[email protected]]# jps

19495 Jps

18849 NameNode

19051 SecondaryNameNode

19228 ResourceManager

[root@masternode centos]# jps

11664 Jps

11418 DataNode

11519 NodeManager

2）必须要关闭防火墙

[root@masternode hadoop-2.7.2]# service iptables stop

http://125.208.30.88:50070/dfshealth.html#tab-datanode

3）禁用selinux

编辑 "/etc/selinux/config"文件，设置"SELINUX=disabled"

或者setenforce 0

10.网页查看集群

（1）访问"http://10.10.10.3.88:50070" 或者http://10.10.10.4:50070

Hadoop安装完成之后，会有两个web管理界面，可以分别通过下面的url查看。在浏览器中输入http://10.10.10.4:8088/，网址为master的ip：

在浏览器中输入：http://10.10.10.4:50070/，网址为master结点所对应的IP：

http://hadoop.apache.org/docs/r2.7.3/hadoop-project-dist/hadoop-common/ClusterSetup.html

11.HDFS的使用

11.1 帮助

如果配置好了环境变量，直接运行hdfs命令，不加任何参数就能得到hdfs的帮助。也可以用hdfs –help得到HDFS的帮助。

[root@masternode hadoop]# hdfs -help

Usage: hdfs [--config confdir] [--loglevel loglevel] COMMAND

where COMMAND is one of:

dfs run a filesystem command on the file systems supported in Hadoop.

classpath prints the classpath

从上面可以看到，dfs命令是执行一个HDFS文件系统支持的命令。可以进一步查看dfs的帮助。下面列出的是hdfs的dfs命令支持的一些文件系统操作，实际上，具体的各个操作的格式可以也从下面看到具体的说明。

[root@masternode hadoop]# hdfs dfs -help

hadoop dfsadmin -report #查看DataNode节点信息，可以使用这个命令脚本监控DFS状况

# hadoop fs -ls hdfs://hcluster:9000/ #指定HDFS地址访问

# hadoop fs -ls / #列出HDFS文件系统目录下文件和目录

# hadoop fs -lsr / #递归列出目录

# hadoop fs -mkdir /test #创建test目录

# hadoop fs -put /root/test.txt /test/test.txt #上传文件到test目录

# hadoop fs -cat /test/test.txt #查看文件内容

# hadoop fs -du /test/test.txt #查看文件大小

# hadoop fs -rm /test/test.txt #删除文件

# hadoop fs -rmr /test #递归删除目录或文件

下面关于Hadoop的文章您也可能喜欢，不妨看看：

ssh background

安科网

Hadoop2.7.1 HA集群部署

零點

Hadoop集群之zookeeper安装配置-hbase安装就不用安装这个模块

安装hadoop

零點

相关推荐

Navicat如何远程连接云服务器数据库

用docker运行postgreSQL的方法步骤

Linux ssh服务器配置代码实例

我用过的几款SSH客户端工具

SSH只能用于远程Linux主机？那说明你见识太小了！

SSH只能用于远程Linux主机？那说明你见识太小了！

学会这5招，让Linux排障更简单

win10下用ssh做端口转发时因为localhost解析引起的失败

使用VSCode的Remote-SSH连接Linux进行远程开发

解决VScode配置远程调试Linux程序的问题

CodeReview实践-Gerrit自动触发JenkinsCI

Git配置及使用(使用云服务器）

Ubuntu 20.04 Install SSH, Change SSH Port, Enable root

ssh免密登陆相关配置

ssh无法启动 (code=exited, status=255)

记录：部署Ansible，Ansible ad-hoc应用(1

ssh配置、vscode使用及常用扩展

华为华三思科交换机设置SSH登录

Ansible自动化

gitlab-Runner配置参数详解

零點