hadoop安装、使用过程的异常汇总

 

2017年06月06日 17:27:39

阅读数：711

1、 启动yarn,resourceManger启动失败，日志报错： 
org.apache.hadoop.yarn.exceptions.YarnRuntimeException: java.net.BindException: Problem binding to [jx-vm04:8031] java.net.BindException: Cannot assign requested address; For more details see: http://wiki.apache.org/hadoop/BindException

定位：
1、netstat –ano | grep '8031'  查不到端口占用；
2、可能是yarn-site.xml配置异常，查看配置信息：
<configuration>

<!-- Site specific YARN configuration properties -->
<!-- 1、指定YARN的老大（ResourceManager）的地址 -->
    <property>
                        <name>yarn.resourcemanager.hostname</name>
                        <value>jx-vm04</value>

    </property>
                <!-- 2、reducer获取数据的方式 -->
    <property>
                        <name>yarn.nodemanager.aux-services</name>
                        <value>mapreduce_shuffle</value>
     </property>


</configuration>
信息配置正常；
3、查看主机配置文件：
$ cat /etc/hosts
127.0.0.1   localhost
#127.0.1.1  jx-vm04
10.10.75.15 jx-vm04

192.168.120.159   mc.zk.thinkjoy.cn
#10.10.75.16 jx-vm01
192.168.3.16 sd-hjy-vm02

本地主机名配置成映射的外网IP（10.10.75.15），这里要使用内网IP地址才行
改成：  127.0.1.1  jx-vm04
      或者   192.168.3.15  jx-vm04

改完保存，重启 start-yarn.sh
查看结果：
/hadoop/hadoop-2.4.1/etc/hadoop$ jps
14338 NameNode
25379 NodeManager
25901 ResourceManager
14468 DataNode
26302 Jps
14666 SecondaryNameNode
启动成功。

 

2、配置单机版hbase，启动后报错：

hadoop@jx-vm04:/usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin$ ./start-hbase.sh 
starting master, logging to /usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin/../logs/hbase-hadoop-master-jx-vm04.out
/usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin/hbase-daemon.sh: 行 189: /usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin/../logs/hbase-hadoop-master-jx-vm04.out: 权限不够
head: 无法打开"/usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin/../logs/hbase-hadoop-master-jx-vm04.out" 读取数据: 没有那个文件或目录
hadoop@jx-vm04:/usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin$ sudo ./start-hbase.sh 
starting master, logging to /usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin/../logs/hbase-root-master-jx-vm04.out
Could not start ZK at requested port of 2181.  ZK was started at port: -1.  Aborting as clients (e.g. shell) will not be able to find this ZK quorum.

 

可以看到在默认账户下提示权限不够，说明要使用管理员账号或者对应安装的权限账号，因为hbase是在hadoop账户下安装的，而zk是在默认的ubuntu账户下安装的，2181端口被ubuntu下的zk占用，而hbase无权访问。

解决办法： 
将sudo下的zk服务占用进程杀掉，在hadoop账户下启动zk： 
hadoop@jx-vm04:/usr/local/wangpei/cloud/zookeeper/zookeeper-3.4.6/bin$ ./zkServer.sh 
JMX enabled by default 
启动hbase： 
./start-hbase.sh 
starting master, logging to /usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin/../logs/hbase-root-master-jx-vm04.out

hbase单机启动成功：

hadoop@jx-vm04:/usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin$ ps -ef | grep hbase
root     11044     1  0 10:56 pts/3    00:00:00 bash /usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin/hbase-daemon.sh --config /usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin/../conf foreground_start master
root     11058 11044  6 10:56 pts/3    00:00:13 /usr/java/jdk1.6.0_45//bin/java -Dproc_master -XX:OnOutOfMemoryError=kill -9 %p -Xmx1000m -XX:+UseConcMarkSweepGC -Dhbase.log.dir=/usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin/../logs -Dhbase.log.file=hbase-root-master-jx-vm04.log -Dhbase.home.dir=/usr/local/wangpei/cloud/hbase/hbase-0.98.24-hadoop2/bin/.. -Dhbase.id.str=root -Dhbase.root.logger=INFO,RFA -Dhbase.security.logger=INFO,RFAS org.apache.hadoop.hbase.master.HMaster start
hadoop   11344 10054  0 10:59 pts/3    00:00:00 grep hbase

• 1
• 2
• 3
• 4

文章标签： hadoopyarnhbase