[求助]启动slaves节点的NodeManager无法启动

搭建了三个节点的集群:
node000: NameNode,DataNode
node001:SecondaryNameNode,DataNode
node002:DataNode

在node000上执行start-dfs.sh启动正常，再执行start-yarn.sh后，发现只有node000上启动了yarn的NodeManager，node001和node002没有启动NodeManager。
查看NodeManager的日志，发现报错：
[mw_shl_code=java,true]2016-01-18 21:58:01,586 INFO org.apache.hadoop.yarn.server.nodemanager.containermanager.localize
r.ResourceLocalizationService: Localizer started on port 10200
2016-01-18 21:58:01,602 INFO org.apache.hadoop.mapred.IndexCache: IndexCache created with max me
mory = 10485760
2016-01-18 21:58:01,619 INFO org.apache.hadoop.service.AbstractService: Service httpshuffle fail
ed in state STARTED; cause: org.jboss.netty.channel.ChannelException: Failed to bind to: 0.0.0.0
/0.0.0.0:13562
org.jboss.netty.channel.ChannelException: Failed to bind to: 0.0.0.0/0.0.0.0:13562
        at org.jboss.netty.bootstrap.ServerBootstrap.bind(ServerBootstrap.java:272)
        at org.apache.hadoop.mapred.ShuffleHandler.serviceStart(ShuffleHandler.java:383)
        at org.apache.hadoop.service.AbstractService.start(AbstractService.java:193)
        at org.apache.hadoop.yarn.server.nodemanager.containermanager.AuxServices.serviceStart(A
uxServices.java:159)
        at org.apache.hadoop.service.AbstractService.start(AbstractService.java:193)
        at org.apache.hadoop.service.CompositeService.serviceStart(CompositeService.java:120)
        at org.apache.hadoop.yarn.server.nodemanager.containermanager.ContainerManagerImpl.servi
ceStart(ContainerManagerImpl.java:307)
        at org.apache.hadoop.service.AbstractService.start(AbstractService.java:193)
        at org.apache.hadoop.service.CompositeService.serviceStart(CompositeService.java:120)
        at org.apache.hadoop.yarn.server.nodemanager.NodeManager.serviceStart(NodeManager.java:2
45)
        at org.apache.hadoop.service.AbstractService.start(AbstractService.java:193)
        at org.apache.hadoop.yarn.server.nodemanager.NodeManager.initAndStartNodeManager(NodeMan
ager.java:426)
        at org.apache.hadoop.yarn.server.nodemanager.NodeManager.main(NodeManager.java:472)
Caused by: java.net.BindException: Address already in use
        at sun.nio.ch.Net.bind0(Native Method)
        at sun.nio.ch.Net.bind(Net.java:463)
        at sun.nio.ch.Net.bind(Net.java:455)
        at sun.nio.ch.ServerSocketChannelImpl.bind(ServerSocketChannelImpl.java:223)
        at sun.nio.ch.ServerSocketAdaptor.bind(ServerSocketAdaptor.java:74)
        at org.jboss.netty.channel.socket.nio.NioServerBoss$RegisterTask.run(NioServerBoss.java:
193)
        at org.jboss.netty.channel.socket.nio.AbstractNioSelector.processTaskQueue(AbstractNioSe
lector.java:366)
        at org.jboss.netty.channel.socket.nio.AbstractNioSelector.run(AbstractNioSelector.java:2
90)
        at org.jboss.netty.channel.socket.nio.NioServerBoss.run(NioServerBoss.java:42)
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
        at java.lang.Thread.run(Thread.java:745)[/mw_shl_code]

配置文件如下：
yarn-site.xml
<configuration>
<!-- Site specific YARN configuration properties -->
    <property>
        <name>yarn.nodemanager.aux-services</name>
        <value>mapreduce_shuffle</value>
    </property>
    <property>
        <name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name>
    <value>org.apache.hadoop.mapred.ShuffleHandler</value>
    </property>
    <property>
        <name>yarn.resourcemanager.scheduler.address</name>
        <value>node000:8030</value>
    </property>
    <property>
        <name>yarn.resourcemanager.resource-tracker.address</name>
        <value>node000:8031</value>
    </property>
    <property>
        <name>yarn.resourcemanager.address</name>
        <value>node000:8032</value>
    </property>
    <property>
        <name>yarn.resourcemanager.admin.address</name>
        <value>node000:8033</value>
    </property>
    <property>
        <name>yarn.resourcemanager.webapp.address</name>
        <value>node000:8088</value>
    </property>
    <property>
        <name>yarn.nodemanager.localizer.address</name>
        <value>node002:10200</value>
    </property>

</configuration>

mapred-site.xml
<configuration>
<property>
        <name>mapreduce.framework.name</name>
        <value>yarn</value>
    </property>
    <property>
        <name>mapreduce.jobhistory.address</name>
        <value>node000:10020</value>
    </property>
    <property>
        <name>mapreduce.jobhistory.webapp.address</name>
        <value>node000:19888</value>
    </property>
</configuration>


请问大神们这个错误是什么原因？

在hosts里面替换下面：
用0.0.0.0代替，不能使用127.0.0.1，

when30 发表于 2016-1-19 11:06
在hosts里面替换下面：
用0.0.0.0代替，不能使用127.0.0.1，

谢谢！
问题解决了。是因为节点上已经启动了yarn的进程，所以13562端口被占用了。