storm on yarn安装时 提交到yarn失败 failed

最近在部署storm on yarn ，部署参考文章

http://www.tuicool.com/articles/BFr2Yv
http://blog.csdn.net/jiushuai/article/details/18729367
 
在安装完zookeeper，配置好storm 和storm on yarn后，启动zookeeper，其中zookeeper的port为2181,
然后通过mvn package 编译工程，发现会出现错误，然后使用mvn packet -DskipTests 重新编译，跳过test
然后向yarn 提交storm任务，storm-yarn launch <path to your storm.yaml file>
 
提交后查看localhost:8088,发现任务failed，查看错误信息发现错误如下

Application application_1411179375629_0005 failed  times due to AM Container for appattempt_1411179375629_0005_000002 exited with exitCode:  due to: Exception from container-launch: org.apache.hadoop.util.Shell$ExitCodeException:
org.apache.hadoop.util.Shell$ExitCodeException:
at org.apache.hadoop.util.Shell.runCommand(Shell.java:)
at org.apache.hadoop.util.Shell.run(Shell.java:)
at org.apache.hadoop.util.Shell$ShellCommandExecutor.execute(Shell.java:)
at org.apache.hadoop.yarn.server.nodemanager.DefaultContainerExecutor.launchContainer(DefaultContainerExecutor.java:)
at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:)
at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:)
at java.util.concurrent.FutureTask.run(FutureTask.java:)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:)
at java.lang.Thread.run(Thread.java:)
Container exited with a non-zero exit code
.Failing this attempt.. Failing the application. 

查看log信息发现具体错误如下

// :: INFO yarn.MasterServer: Starting Master Thrift Server
// :: ERROR auth.ThriftServer: ThriftServer is being stopped due to: org.apache.thrift7.transport.TTransportException: Could not create ServerSocket on address 0.0.0.0/0.0.0.0:.
org.apache.thrift7.transport.TTransportException: Could not create ServerSocket on address 0.0.0.0/0.0.0.0:.
        at org.apache.thrift7.transport.TNonblockingServerSocket.<init>(TNonblockingServerSocket.java:)
        at org.apache.thrift7.transport.TNonblockingServerSocket.<init>(TNonblockingServerSocket.java:)
        at org.apache.thrift7.transport.TNonblockingServerSocket.<init>(TNonblockingServerSocket.java:)
        at backtype.storm.security.auth.SimpleTransportPlugin.getServer(SimpleTransportPlugin.java:)
        at backtype.storm.security.auth.ThriftServer.serve(ThriftServer.java:)
        at com.yahoo.storm.yarn.MasterServer.main(MasterServer.java:)

在启动Master Thrift Server时发生错误，错误为端口错误，因为9000端口已经被hdfs占用监听，因此不能创建ServerSocket

致谢google，搜到的解决办法为修改端口号，帖子如下

https://groups.google.com/forum/#!topic/storm-yarn/A1ds1M6qmN8

修改storm-yarn-master/src/main/resources/master_defaults.yaml，将其中的master.thrift.port修改为一个其他的合适的值，我修改为9001

然后再编译工程，重新提交，这时发现任务没有failed，但是访问localhost:7070，不能访问，查看log后发现错误为nimbus没有启动成功，错误如下：

// :: ERROR yarn.MasterServer: Unhandled error in AM:
org.apache.hadoop.yarn.exceptions.InvalidResourceRequestException: Invalid resource request, requested virtual cores < , or requested virtual cores > max configured, requestedVirtualCores=, maxVirtualCores=
    at org.apache.hadoop.yarn.server.resourcemanager.scheduler.SchedulerUtils.validateResourceRequest(SchedulerUtils.java:)
    at org.apache.hadoop.yarn.server.resourcemanager.RMServerUtils.validateResourceRequests(RMServerUtils.java:)
    at org.apache.hadoop.yarn.server.resourcemanager.ApplicationMasterService.allocate(ApplicationMasterService.java:)
    ... ... ... ... ... ...
    at com.yahoo.storm.yarn.MasterServer$.run(MasterServer.java:)
Caused by: org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.yarn.exceptions.InvalidResourceRequestException): Invalid resource request, requested virtual cores < , or requested virtual cores > max configured, requestedVirtualCores=, maxVirtualCores=
    at org.apache.hadoop.yarn.server.resourcemanager.scheduler.SchedulerUtils.validateResourceRequest(SchedulerUtils.java:)
    at org.apache.hadoop.yarn.server.resourcemanager.RMServerUtils.validateResourceRequests(RMServerUtils.java:)

    ... ... ... ... ... ...

 at com.sun.proxy.$Proxy7.allocate(Unknown Source) at org.apache.hadoop.yarn.api.impl.pb.client.ApplicationMasterProtocolPBClientImpl.allocate(ApplicationMasterProtocolPBClientImpl.java:) ...  more // :: INFO yarn.StormMasterServerHandler: stopping supervisors... // :: INFO yarn.StormMasterServerHandler: stopping UI... // :: INFO yarn.StormMasterServerHandler: stopping nimbus...

意思是申请的virtual cores超过最大限制maxVirtualCores，google后找到帖子

hadoop - Why cannot more than 32 cores be requested from YARN to run a job? - Stack Overflow
http://stackoverflow.com/questions/29780401/why-cannot-more-than-32-cores-be-requested-from-yarn-to-run-a-job

因此修改yarn-site.xml如下

<property>
  <name>yarn.nodemanager.resource.cpu-vcores</name>
  <value></value>
</property>

重新启动yarn，然后提交，成功！！！

如果发现以下错误

-- :: o.a.z.ClientCnxn [INFO] Opening socket connection to server MMC/192.168.1.200:
-- :: o.a.z.ClientCnxn [WARN] Session 0x0 for server null, unexpected error, closing socket connection and attempting reconnect
java.net.ConnectException: 拒绝连接

有可能是zookeeper没有启动，启动即可

时刻注意自己的防火墙有没有关闭，有些不知名的原因是因为防火墙没有关闭造成的！

用了三四天时间才解决了这个问题，期间走了不少弯路，因为刚开始接触hadoop storm on yarn，因此有时候出错误了不知道怎么去查错，开始时只是自己估计是哪出了问题，然后改一下重新跑一下，结果还是不行，后来学会了去查错误log，在masterhost：8088/logs/下就是所有的错误日志，然后查找相关的错误日志排错事半功倍

致谢：google

吐槽一下万恶的网G络F审W查￥制#度，我操我操我操我操我操！！！Internet上的闭关锁国！！！