Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
MapReduce, mail # user - Re: Permission related errors when running with a different user


Copy link to this message
-
Re: Permission related errors when running with a different user
Krishna Kishore Bonagiri 2012-12-07, 10:47
Hi Harsh,

 Thanks for the quick reply. I have tried this but it didn't work exactly.
Upon your suggestion I re-looked into the error and thought the following
might work

hadoop fs -chmod 777 /

I tried it and it worked.

  I went past that error. I now have a different error. I am seeing this
error in the Resource Manager's logs. Can you please throw me some clue on
this too..

2012-12-07 05:34:22,421 INFO  fifo.FifoScheduler
(FifoScheduler.java:containerCompleted(721)) - Application
appattempt_1353856203101_0369_000001 released container
container_1353856203101_0369_01_000003 on node: host: isredeng:51271
#containers=1 available=8064 used=128 with event: FINISHED
2012-12-07 05:34:24,401 WARN  attempt.RMAppAttemptImpl
(RMAppAttemptImpl.java:generateProxyUriWithoutScheme(379)) - Could not
proxify
java.net.URISyntaxException: Expected authority at index 7: http://
        at java.net.URI$Parser.fail(URI.java:2820)
        at java.net.URI$Parser.failExpecting(URI.java:2826)
        at java.net.URI$Parser.parseHierarchical(URI.java:3065)
        at java.net.URI$Parser.parse(URI.java:3025)
        at java.net.URI.<init>(URI.java:589)
        at
org.apache.hadoop.yarn.server.webproxy.ProxyUriUtils.getUriFromAMUrl(ProxyUriUtils.java:143)
        at
org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl.generateProxyUriWithoutScheme(RMAppAttemptImpl.java:371)
        at
org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl.access$2500(RMAppAttemptImpl.java:81)
        at
org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl$AMUnregisteredTransition.transition(RMAppAttemptImpl.java:849)
        at
org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl$AMUnregisteredTransition.transition(RMAppAttemptImpl.java:835)
        at
org.apache.hadoop.yarn.state.StateMachineFactory$SingleInternalArc.doTransition(StateMachineFactory.java:357)
        at
org.apache.hadoop.yarn.state.StateMachineFactory.doTransition(StateMachineFactory.java:298)
        at
org.apache.hadoop.yarn.state.StateMachineFactory.access$300(StateMachineFactory.java:43)
        at
org.apache.hadoop.yarn.state.StateMachineFactory$InternalStateMachine.doTransition(StateMachineFactory.java:443)
        at
org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl.handle(RMAppAttemptImpl.java:476)
        at
org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl.handle(RMAppAttemptImpl.java:80)
        at
org.apache.hadoop.yarn.server.resourcemanager.ResourceManager$ApplicationAttemptEventDispatcher.handle(ResourceManager.java:414)
        at
org.apache.hadoop.yarn.server.resourcemanager.ResourceManager$ApplicationAttemptEventDispatcher.handle(ResourceManager.java:395)
        at
org.apache.hadoop.yarn.event.AsyncDispatcher.dispatch(AsyncDispatcher.java:125)
        at
org.apache.hadoop.yarn.event.AsyncDispatcher$1.run(AsyncDispatcher.java:74)
        at java.lang.Thread.run(Thread.java:736)
Thanks,
Kishore

On Thu, Dec 6, 2012 at 8:58 PM, Harsh J <[EMAIL PROTECTED]> wrote:

> You are attempting a job submit operation as user "root" over HDFS and
> MR. Running a job involves placing the requisite files on HDFS, so MR
> can leverage its distributed presence to run task work.
>
> The files are usually placed under a user's HDFS home directory, which
> is of the form /user/[NAME]. By default, HDFS has no notions of a user
> existing in it (Imagine a linux user account with no home-dir yet). So
> you'll first have to provision the user with a home directory he can
> own for himself, as the HDFS administrator:
>
> Create and grant ownership to the user root, his own homedir:
>
> sudo -u hdfs hadoop fs -mkdir -p /user/root/
> sudo -u hdfs hadoop fs -chown root:root /user/root
>
> Once this is done, you can try to resubmit the job and the
> AccessControlException should be resolved.
>
> On Thu, Dec 6, 2012 at 8:28 PM, Krishna Kishore Bonagiri
> <[EMAIL PROTECTED]> wrote: