I am bringing up a Hadoop cluster for the first time (but am an experienced sysadmin with lots of cluster experience) and running into an issue with permissions on mapred.system.dir.   It has generally been a chore to figure out all the various directories that need to be created to get Hadoop working, some on the local FS, others within HDFS, getting the right ownership and permissions, etc..  I think I am mostly there but can't seem to get past my current issue with mapred.system.dir.

Some general info first:
OS: RHEL6
Hadoop version: hadoop-1.0.3-1.x86_64

20 node cluster configured as follows
1 node as primary namenode
1 node as secondary namenode + job tracker
18 nodes as datanode + tasktracker

I have HDFS up and running and have the following in mapred-site.xml:
<property>
  <name>mapred.system.dir</name>
  <value>hdfs://hadoop1/mapred</value>
  <description>Shared data for JT - this must be in HDFS</description>
</property>

I have created this directory in HDFS, owner mapred:hadoop, permissions 700 which seems to be the most common recommendation amongst multiple, often conflicting articles about how to set up Hadoop.  Here is the top level of my filesystem:
hyperion-hdp4@hdfs:hadoop fs -ls /
Found 3 items
drwx------   - mapred hadoop          0 2012-10-09 12:58 /mapred
drwxrwxrwx   - hdfs   hadoop          0 2012-10-09 13:00 /tmp
drwxr-xr-x   - hdfs   hadoop          0 2012-10-09 12:51 /user

Note, it doesn't seem to really matter what permissions I set on /mapred since when the Jobtracker starts up it changes them to 700.

However, when I try to run the hadoop example teragen program as a "regular" user I am getting this error:
hyperion-hdp4@robing:hadoop jar /usr/share/hadoop/hadoop-examples*.jar teragen -D dfs.block.size=536870912 10000000000 /user/robing/terasort-input
Generating 10000000000 using 2 maps with step of 5000000000
12/10/09 16:27:02 INFO mapred.JobClient: Running job: job_201210072045_0003
12/10/09 16:27:03 INFO mapred.JobClient:  map 0% reduce 0%
12/10/09 16:27:03 INFO mapred.JobClient: Job complete: job_201210072045_0003
12/10/09 16:27:03 INFO mapred.JobClient: Counters: 0
12/10/09 16:27:03 INFO mapred.JobClient: Job Failed: Job initialization failed:
org.apache.hadoop.security.AccessControlException: org.apache.hadoop.security.AccessControlException: Permission denied: user=robing, access=EXECUTE, inode="mapred":mapred:hadoop:rwx------
at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:39)
at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:27)
at java.lang.reflect.Constructor.newInstance(Constructor.java:513)
at org.apache.hadoop.ipc.RemoteException.instantiateException(RemoteException.java:95)
at org.apache.hadoop.ipc.RemoteException.unwrapRemoteException(RemoteException.java:57)
at org.apache.hadoop.hdfs.DFSClient$DFSOutputStream.<init>(DFSClient.java:3251)
at org.apache.hadoop.hdfs.DFSClient.create(DFSClient.java:713)
at org.apache.hadoop.hdfs.DistributedFileSystem.create(DistributedFileSystem.java:182)
at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:555)
at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:536)
at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:443)
at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:435)
at org.apache.hadoop.security.Credentials.writeTokenStorageFile(Credentials.java:169)
at org.apache.hadoop.mapred.JobInProgress.generateAndStoreTokens(JobInProgress.java:3537)
at org.apache.hadoop.mapred.JobInProgress.initTasks(JobInProgress.java:696)
at org.apache.hadoop.mapred.JobTracker.initJob(JobTracker.java:4207)
at org.apache.hadoop.mapred.FairScheduler$JobInitializer$InitJob.run(FairScheduler.java:291)
at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)
at java.lang.Thread.run(Thread.java:662)
<rest of stack trace omitted>

This seems to be saying that is trying to write to the HDFS /mapred filesystem as me (robing) rather than as mapred, the username under which the jobtracker and tasktracker run.

To verify this is what is happening, I manually changed the permissions on /mapred from 700 to 755 since it claims to want execute access:
hyperion-hdp4@mapred:hadoop fs -chmod 755 /mapred
hyperion-hdp4@mapred:hadoop fs -ls /
Found 3 items
drwxr-xr-x   - mapred hadoop          0 2012-10-09 12:58 /mapred
drwxrwxrwx   - hdfs   hadoop          0 2012-10-09 13:00 /tmp
drwxr-xr-x   - hdfs   hadoop          0 2012-10-09 12:51 /user
hyperion-hdp4@mapred:

Now I try running again and it fails again, this time complaining it wants write access to /mapred:
hyperion-hdp4@robing:hadoop jar /usr/share/hadoop/hadoop-examples*.jar teragen -D dfs.block.size=536870912 10000000000 /user/robing/terasort-input
Generating 10000000000 using 2 maps with step of 5000000000
12/10/09 16:31:29 INFO mapred.JobClient: Running job: job_201210072045_0005
12/10/09 16:31:30 INFO mapred.JobClient:  map 0% reduce 0%
12/10/09 16:31:30 INFO mapred.JobClient: Job complete: job_201210072045_0005
12/10/09 16:31:30 INFO mapred.JobClient: Counters: 0
12/10/09 16:31:30 INFO mapred.JobClient: Job Failed: Job initialization failed:
org.apache.hadoop.security.AccessControlException: org.apache.hadoop.security.AccessControlException: Permission denied: user=robing, access=WRITE, inode="mapred":mapred:hadoop:rwxr-xr-x

So I changed the permissions on /mapred to 777:
hyperion-hdp4@mapred:hadoop fs -chmod 777 /mapred
hyperion-hdp4@mapred:hadoop fs -ls /
Found 3 items
drwxrwxrwx   - mapred hadoop          0 2012-10-09 12:58 /mapred
drwxrwxrwx   - hdfs   hadoop          0 2012-10-09 13:00 /tmp
drwxr-xr-x   - hdfs   hadoop          0 2012-10-09 12:51 /user
hyperion-hdp4@mapred:

And then I run again and this time it works.
yperion-hdp4@robing:hadoop jar /usr/share/hadoop/hadoop-exa

what is your "mapreduce.jobtracker.staging.root.dir" set to. This is a directory that needs to be writable by the user and is is recommended to be set to "/user" so it writes in appropriate users home directory.

--
Arpit Gupta
Hortonworks Inc.
http://hortonworks.com/

On Oct 9, 2012, at 4:44 PM, "Goldstone, Robin J." <[EMAIL PROTECTED]> wrote:

> I am bringing up a Hadoop cluster for the first time (but am an experienced sysadmin with lots of cluster experience) and running into an issue with permissions on mapred.system.dir.   It has generally been a chore to figure out all the various directories that need to be created to get Hadoop working, some on the local FS, others within HDFS, getting the right ownership and permissions, etc..  I think I am mostly there but can't seem to get past my current issue with mapred.system.dir.
>
> Some general info first:
> OS: RHEL6
> Hadoop version: hadoop-1.0.3-1.x86_64
>
> 20 node cluster configured as follows
> 1 node as primary namenode
> 1 node as secondary namenode + job tracker
> 18 nodes as datanode + tasktracker
>
> I have HDFS up and running and have the following in mapred-site.xml:
> <property>
>   <name>mapred.system.dir</name>
>   <value>hdfs://hadoop1/mapred</value>
>   <description>Shared data for JT - this must be in HDFS</description>
> </property>
>
> I have created this directory in HDFS, owner mapred:hadoop, permissions 700 which seems to be the most common recommendation amongst multiple, often conflicting articles about how to set up Hadoop.  Here is the top level of my filesystem:
> hyperion-hdp4@hdfs:hadoop fs -ls /
> Found 3 items
> drwx------   - mapred hadoop          0 2012-10-09 12:58 /mapred
> drwxrwxrwx   - hdfs   hadoop          0 2012-10-09 13:00 /tmp
> drwxr-xr-x   - hdfs   hadoop          0 2012-10-09 12:51 /user
>
> Note, it doesn't seem to really matter what permissions I set on /mapred since when the Jobtracker starts up it changes them to 700.  
>
> However, when I try to run the hadoop example teragen program as a "regular" user I am getting this error:
> hyperion-hdp4@robing:hadoop jar /usr/share/hadoop/hadoop-examples*.jar teragen -D dfs.block.size=536870912 10000000000 /user/robing/terasort-input
> Generating 10000000000 using 2 maps with step of 5000000000
> 12/10/09 16:27:02 INFO mapred.JobClient: Running job: job_201210072045_0003
> 12/10/09 16:27:03 INFO mapred.JobClient:  map 0% reduce 0%
> 12/10/09 16:27:03 INFO mapred.JobClient: Job complete: job_201210072045_0003
> 12/10/09 16:27:03 INFO mapred.JobClient: Counters: 0
> 12/10/09 16:27:03 INFO mapred.JobClient: Job Failed: Job initialization failed:
> org.apache.hadoop.security.AccessControlException: org.apache.hadoop.security.AccessControlException: Permission denied: user=robing, access=EXECUTE, inode="mapred":mapred:hadoop:rwx------
> at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
> at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:39)
> at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:27)
> at java.lang.reflect.Constructor.newInstance(Constructor.java:513)
> at org.apache.hadoop.ipc.RemoteException.instantiateException(RemoteException.java:95)
> at org.apache.hadoop.ipc.RemoteException.unwrapRemoteException(RemoteException.java:57)
> at org.apache.hadoop.hdfs.DFSClient$DFSOutputStream.<init>(DFSClient.java:3251)
> at org.apache.hadoop.hdfs.DFSClient.create(DFSClient.java:713)
> at org.apache.hadoop.hdfs.DistributedFileSystem.create(DistributedFileSystem.java:182)
> at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:555)
> at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:536)
> at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:443)
> at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:435)
> at org.apache.hadoop.security.Credentials.writeTokenStorageFile(Credentials.java:169)
> at org.apache.hadoop.mapred.JobInProgress.generateAndStoreTokens(JobInProgress.java:3537)

What is the permission for /hadoop1 dir in HDFS? Is "mapred" user have
permission on the same directory?

Thanks,
Manu S

On Wed, Oct 10, 2012 at 5:52 AM, Arpit Gupta <[EMAIL PROTECTED]> wrote:

> what is your "mapreduce.jobtracker.staging.root.dir" set to. This is a
> directory that needs to be writable by the user and is is recommended to be
> set to "/user" so it writes in appropriate users home directory.
>
> --
> Arpit Gupta
> Hortonworks Inc.
> http://hortonworks.com/
> <http://www.hadoopsummit.org/>
>
> On Oct 9, 2012, at 4:44 PM, "Goldstone, Robin J." <[EMAIL PROTECTED]>
> wrote:
>
>  I am bringing up a Hadoop cluster for the first time (but am an
> experienced sysadmin with lots of cluster experience) and running into an
> issue with permissions on mapred.system.dir.   It has generally been a
> chore to figure out all the various directories that need to be created to
> get Hadoop working, some on the local FS, others within HDFS, getting the
> right ownership and permissions, etc..  I think I am mostly there but can't
> seem to get past my current issue with mapred.system.dir.
>
>  Some general info first:
> OS: RHEL6
> Hadoop version: hadoop-1.0.3-1.x86_64
>
>  20 node cluster configured as follows
> 1 node as primary namenode
> 1 node as secondary namenode + job tracker
> 18 nodes as datanode + tasktracker
>
>  I have HDFS up and running and have the following in mapred-site.xml:
>  <property>
>   <name>mapred.system.dir</name>
>   <value>hdfs://hadoop1/mapred</value>
>   <description>Shared data for JT - this must be in HDFS</description>
> </property>
>
>  I have created this directory in HDFS, owner mapred:hadoop, permissions
> 700 which seems to be the most common recommendation amongst multiple,
> often conflicting articles about how to set up Hadoop.  Here is the top
> level of my filesystem:
>  hyperion-hdp4@hdfs:hadoop fs -ls /
> Found 3 items
> drwx------   - mapred hadoop          0 2012-10-09 12:58 /mapred
> drwxrwxrwx   - hdfs   hadoop          0 2012-10-09 13:00 /tmp
> drwxr-xr-x   - hdfs   hadoop          0 2012-10-09 12:51 /user
>
>  Note, it doesn't seem to really matter what permissions I set on /mapred
> since when the Jobtracker starts up it changes them to 700.
>
>  However, when I try to run the hadoop example teragen program as a
> "regular" user I am getting this error:
>  hyperion-hdp4@robing:hadoop jar /usr/share/hadoop/hadoop-examples*.jar
> teragen -D dfs.block.size=536870912 10000000000 /user/robing/terasort-input
> Generating 10000000000 using 2 maps with step of 5000000000
> 12/10/09 16:27:02 INFO mapred.JobClient: Running job: job_201210072045_0003
> 12/10/09 16:27:03 INFO mapred.JobClient:  map 0% reduce 0%
> 12/10/09 16:27:03 INFO mapred.JobClient: Job complete:
> job_201210072045_0003
> 12/10/09 16:27:03 INFO mapred.JobClient: Counters: 0
> 12/10/09 16:27:03 INFO mapred.JobClient: Job Failed: Job initialization
> failed:
> org.apache.hadoop.security.AccessControlException:
> org.apache.hadoop.security.AccessControlException: Permission denied:
> user=robing, access=EXECUTE, inode="mapred":mapred:hadoop:rwx------
> at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
> at
> sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:39)
> at
> sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:27)
> at java.lang.reflect.Constructor.newInstance(Constructor.java:513)
> at
> org.apache.hadoop.ipc.RemoteException.instantiateException(RemoteException.java:95)
> at
> org.apache.hadoop.ipc.RemoteException.unwrapRemoteException(RemoteException.java:57)
> at
> org.apache.hadoop.hdfs.DFSClient$DFSOutputStream.<init>(DFSClient.java:3251)
> at org.apache.hadoop.hdfs.DFSClient.create(DFSClient.java:713)
> at
> org.apache.hadoop.hdfs.DistributedFileSystem.create(DistributedFileSystem.java:182)
> at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:555)
> at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:536)

There is no /hadoop1 directory.  It is //hadoop1 which is the name of the server running the name node daemon:

<value>hdfs://hadoop1/mapred</value>
Per offline conversations with Arpit, it appears this problem is related to the fact that I am using the fair scheduler.  The fair scheduler is designed to run map reduce jobs as the user, rather than under the mapred username.  Apparently there are some issues with this scheduler related to permissions on certain directories not allowing other users to execute/write in places that are necessary for the job to run.  I haven't yet tried Arpit's suggestion to switch to the task scheduler but I imagine it will resolve my issue, at least for now.  Ultimately I do want to use the fair scheduler, as multi-tenancy is a key requirement for our Hadoop deployment.

From: Manu S <[EMAIL PROTECTED]<mailto:[EMAIL PROTECTED]>>
Reply-To: "[EMAIL PROTECTED]<mailto:[EMAIL PROTECTED]>" <[EMAIL PROTECTED]<mailto:[EMAIL PROTECTED]>>
Date: Wednesday, October 10, 2012 3:34 AM
To: "[EMAIL PROTECTED]<mailto:[EMAIL PROTECTED]>" <[EMAIL PROTECTED]<mailto:[EMAIL PROTECTED]>>
Subject: Re: issue with permissions of mapred.system.dir

What is the permission for /hadoop1 dir in HDFS? Is "mapred" user have permission on the same directory?

Thanks,
Manu S

On Wed, Oct 10, 2012 at 5:52 AM, Arpit Gupta <[EMAIL PROTECTED]<mailto:[EMAIL PROTECTED]>> wrote:
what is your "mapreduce.jobtracker.staging.root.dir" set to. This is a directory that needs to be writable by the user and is is recommended to be set to "/user" so it writes in appropriate users home directory.

--
Arpit Gupta
Hortonworks Inc.
http://hortonworks.com/
<http://www.hadoopsummit.org/>

On Oct 9, 2012, at 4:44 PM, "Goldstone, Robin J." <[EMAIL PROTECTED]<mailto:[EMAIL PROTECTED]>> wrote:

I am bringing up a Hadoop cluster for the first time (but am an experienced sysadmin with lots of cluster experience) and running into an issue with permissions on mapred.system.dir.   It has generally been a chore to figure out all the various directories that need to be created to get Hadoop working, some on the local FS, others within HDFS, getting the right ownership and permissions, etc..  I think I am mostly there but can't seem to get past my current issue with mapred.system.dir.

Some general info first:
OS: RHEL6
Hadoop version: hadoop-1.0.3-1.x86_64

20 node cluster configured as follows
1 node as primary namenode
1 node as secondary namenode + job tracker
18 nodes as datanode + tasktracker

I have HDFS up and running and have the following in mapred-site.xml:
<property>
  <name>mapred.system.dir</name>
  <value>hdfs://hadoop1/mapred</value>
  <description>Shared data for JT - this must be in HDFS</description>
</property>

I have created this directory in HDFS, owner mapred:hadoop, permissions 700 which seems to be the most common recommendation amongst multiple, often conflicting articles about how to set up Hadoop.  Here is the top level of my filesystem:
hyperion-hdp4@hdfs:hadoop fs -ls /
Found 3 items
drwx------   - mapred hadoop          0 2012-10-09 12:58 /mapred
drwxrwxrwx   - hdfs   hadoop          0 2012-10-09 13:00 /tmp
drwxr-xr-x   - hdfs   hadoop          0 2012-10-09 12:51 /user

Note, it doesn't seem to really matter what permissions I set on /mapred since when the Jobtracker starts up it changes them to 700.

However, when I try to run the hadoop example teragen program as a "regular" user I am getting this error:
hyperion-hdp4@robing:hadoop jar /usr/share/hadoop/hadoop-examples*.jar teragen -D dfs.block.size=536870912 10000000000 /user/robing/terasort-input
Generating 10000000000 using 2 maps with step of 5000000000
12/10/09 16:27:02 INFO mapred.JobClient: Running job: job_201210072045_0003
12/10/09 16:27:03 INFO mapred.JobClient:  map 0% reduce 0%
12/10/09 16:27:03 INFO mapred.JobClient: Job complete: job_201210072045_0003
12/10/09 16:27:03 INFO mapred.JobClient: Counters: 0
12/10/09 16:27:03 INFO mapred.JobClient: Job Failed: Job initialization failed:
org.apache.hadoop.security.AccessControlException: org.apache.hadoop.security.AccessControlException: Permission denied: user=robing, access=EXECUTE, inode="mapred":mapred:hadoop:rwx------
at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:39)
at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:27)
at java.lang.reflect.Constructor.newInstance(Constructor.java:513)
at org.apache.hadoop.ipc.RemoteException.instantiateException(RemoteException.java:95)
at org.apache.hadoop.ipc.RemoteException.unwrapRemoteException(RemoteException.java:57)
at org.apache.hadoop.hdfs.DFSClient$DFSOutputStream.<init>(DFSClient.java:3251)
at org.apache.hadoop.hdfs.DFSClient.create(DFSClient.java:713)
at org.apache.hadoop.hdfs.DistributedFileSystem.create(DistributedFileSystem.java:182)
at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:555)
at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:536)
at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:443)
at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:435)
at org.apache.hadoop.security.Credentials.writeTokenStorageFile(Credentials.java:169)
at org.apache.hadoop.mapred.JobInProgress.generateAndStoreTokens(JobInProgress.java:3537)
at org.apache.hadoop.mapred.JobInProgress.initTasks(JobInProgress.java:696)
at org.apache.hadoop.mapred.JobTracker.initJob(JobTracker.java:4207)
at org.apache.hadoop.mapred.FairScheduler$JobInitializer$InitJob.run(FairScheduler.java:291)
at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)
at java.lang.Thread.run(Thread.java:662)
<rest of stack trace omitted>

This seems to be saying that is trying to write to the HDFS /mapred filesystem as me (r

Robin

I will try to investigate the issue with fair scheduler. Do let us know if switching to default or the capacity scheduler solved the issue.

--
Arpit Gupta
Hortonworks Inc.
http://hortonworks.com/

On Oct 10, 2012, at 9:32 AM, "Goldstone, Robin J." <[EMAIL PROTECTED]> wrote:

> There is no /hadoop1 directory.  It is //hadoop1 which is the name of the server running the name node daemon:
>
>> <value>hdfs://hadoop1/mapred</value>
>
> Per offline conversations with Arpit, it appears this problem is related to the fact that I am using the fair scheduler.  The fair scheduler is designed to run map reduce jobs as the user, rather than under the mapred username.  Apparently there are some issues with this scheduler related to permissions on certain directories not allowing other users to execute/write in places that are necessary for the job to run.  I haven't yet tried Arpit's suggestion to switch to the task scheduler but I imagine it will resolve my issue, at least for now.  Ultimately I do want to use the fair scheduler, as multi-tenancy is a key requirement for our Hadoop deployment.
>
> From: Manu S <[EMAIL PROTECTED]>
> Reply-To: "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>
> Date: Wednesday, October 10, 2012 3:34 AM
> To: "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>
> Subject: Re: issue with permissions of mapred.system.dir
>
> What is the permission for /hadoop1 dir in HDFS? Is "mapred" user have permission on the same directory?
>
> Thanks,
> Manu S
>
> On Wed, Oct 10, 2012 at 5:52 AM, Arpit Gupta <[EMAIL PROTECTED]> wrote:
>> what is your "mapreduce.jobtracker.staging.root.dir" set to. This is a directory that needs to be writable by the user and is is recommended to be set to "/user" so it writes in appropriate users home directory.
>>
>> --
>> Arpit Gupta
>> Hortonworks Inc.
>> http://hortonworks.com/
>>
>> On Oct 9, 2012, at 4:44 PM, "Goldstone, Robin J." <[EMAIL PROTECTED]> wrote:
>>
>>> I am bringing up a Hadoop cluster for the first time (but am an experienced sysadmin with lots of cluster experience) and running into an issue with permissions on mapred.system.dir.   It has generally been a chore to figure out all the various directories that need to be created to get Hadoop working, some on the local FS, others within HDFS, getting the right ownership and permissions, etc..  I think I am mostly there but can't seem to get past my current issue with mapred.system.dir.
>>>
>>> Some general info first:
>>> OS: RHEL6
>>> Hadoop version: hadoop-1.0.3-1.x86_64
>>>
>>> 20 node cluster configured as follows
>>> 1 node as primary namenode
>>> 1 node as secondary namenode + job tracker
>>> 18 nodes as datanode + tasktracker
>>>
>>> I have HDFS up and running and have the following in mapred-site.xml:
>>> <property>
>>>   <name>mapred.system.dir</name>
>>>   <value>hdfs://hadoop1/mapred</value>
>>>   <description>Shared data for JT - this must be in HDFS</description>
>>> </property>
>>>
>>> I have created this directory in HDFS, owner mapred:hadoop, permissions 700 which seems to be the most common recommendation amongst multiple, often conflicting articles about how to set up Hadoop.  Here is the top level of my filesystem:
>>> hyperion-hdp4@hdfs:hadoop fs -ls /
>>> Found 3 items
>>> drwx------   - mapred hadoop          0 2012-10-09 12:58 /mapred
>>> drwxrwxrwx   - hdfs   hadoop          0 2012-10-09 13:00 /tmp
>>> drwxr-xr-x   - hdfs   hadoop          0 2012-10-09 12:51 /user
>>>
>>> Note, it doesn't seem to really matter what permissions I set on /mapred since when the Jobtracker starts up it changes them to 700.  
>>>
>>> However, when I try to run the hadoop example teragen program as a "regular" user I am getting this error:
>>> hyperion-hdp4@robing:hadoop jar /usr/share/hadoop/hadoop-examples*.jar teragen -D dfs.block.size=536870912 10000000000 /user/robing/terasort-input
>>> Generating 10000000000 using 2 maps with step of 5000000000
>>> 12/10/09 16:27:02 INFO mapred.JobClient: Running job: job_201210072045_0003

It's a known issue for fairscheduler in Hadoop 1.x, see
MAPREDUCE-4398. A workaround is to submit 4 or more jobs as the user
of the jobtracker and everything will work fine afterwards. BTW, IBM
BigInsights community version (open source) contains the right fix
(properly initialize job init threads) since BigInsights 1.3.0.1.
Unfortunately IBM devs are too busy to port/submit the patches to
Apache right now :)

__Luke

On Wed, Oct 10, 2012 at 9:32 AM, Goldstone, Robin J.
<[EMAIL PROTECTED]> wrote:
> There is no /hadoop1 directory.  It is //hadoop1 which is the name of the
> server running the name node daemon:
>
> <value>hdfs://hadoop1/mapred</value>
>
> Per offline conversations with Arpit, it appears this problem is related to
> the fact that I am using the fair scheduler.  The fair scheduler is designed
> to run map reduce jobs as the user, rather than under the mapred username.
> Apparently there are some issues with this scheduler related to permissions
> on certain directories not allowing other users to execute/write in places
> that are necessary for the job to run.  I haven't yet tried Arpit's
> suggestion to switch to the task scheduler but I imagine it will resolve my
> issue, at least for now.  Ultimately I do want to use the fair scheduler, as
> multi-tenancy is a key requirement for our Hadoop deployment.
>
> From: Manu S <[EMAIL PROTECTED]>
> Reply-To: "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>
> Date: Wednesday, October 10, 2012 3:34 AM
> To: "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>
> Subject: Re: issue with permissions of mapred.system.dir
>
> What is the permission for /hadoop1 dir in HDFS? Is "mapred" user have
> permission on the same directory?
>
> Thanks,
> Manu S
>
> On Wed, Oct 10, 2012 at 5:52 AM, Arpit Gupta <[EMAIL PROTECTED]> wrote:
>>
>> what is your "mapreduce.jobtracker.staging.root.dir" set to. This is a
>> directory that needs to be writable by the user and is is recommended to be
>> set to "/user" so it writes in appropriate users home directory.
>>
>> --
>> Arpit Gupta
>> Hortonworks Inc.
>> http://hortonworks.com/
>>
>> On Oct 9, 2012, at 4:44 PM, "Goldstone, Robin J." <[EMAIL PROTECTED]>
>> wrote:
>>
>> I am bringing up a Hadoop cluster for the first time (but am an
>> experienced sysadmin with lots of cluster experience) and running into an
>> issue with permissions on mapred.system.dir.   It has generally been a chore
>> to figure out all the various directories that need to be created to get
>> Hadoop working, some on the local FS, others within HDFS, getting the right
>> ownership and permissions, etc..  I think I am mostly there but can't seem
>> to get past my current issue with mapred.system.dir.
>>
>> Some general info first:
>> OS: RHEL6
>> Hadoop version: hadoop-1.0.3-1.x86_64
>>
>> 20 node cluster configured as follows
>> 1 node as primary namenode
>> 1 node as secondary namenode + job tracker
>> 18 nodes as datanode + tasktracker
>>
>> I have HDFS up and running and have the following in mapred-site.xml:
>> <property>
>>   <name>mapred.system.dir</name>
>>   <value>hdfs://hadoop1/mapred</value>
>>   <description>Shared data for JT - this must be in HDFS</description>
>> </property>
>>
>> I have created this directory in HDFS, owner mapred:hadoop, permissions
>> 700 which seems to be the most common recommendation amongst multiple, often
>> conflicting articles about how to set up Hadoop.  Here is the top level of
>> my filesystem:
>> hyperion-hdp4@hdfs:hadoop fs -ls /
>> Found 3 items
>> drwx------   - mapred hadoop          0 2012-10-09 12:58 /mapred
>> drwxrwxrwx   - hdfs   hadoop          0 2012-10-09 13:00 /tmp
>> drwxr-xr-x   - hdfs   hadoop          0 2012-10-09 12:51 /user
>>
>> Note, it doesn't seem to really matter what permissions I set on /mapred
>> since when the Jobtracker starts up it changes them to 700.
>>
>> However, when I try to run the hadoop example teragen program as a
>> "regular" user I am getting this error:
>> hyperion-hdp4@robing:hadoop jar /usr/share/hadoop/hadoop-examples*.jar

On 10/09/2012 07:44 PM, Goldstone, Robin J. wrote:
> I am bringing up a Hadoop cluster for the first time (but am an
> experienced sysadmin with lots of cluster experience) and running into
> an issue with permissions on mapred.system.dir. It has generally been
> a chore to figure out all the various directories that need to be
> created to get Hadoop working, some on the local FS, others within
> HDFS, getting the right ownership and permissions, etc..  I think I am
> mostly there but can't seem to get past my current issue with
> mapred.system.dir.
>
> Some general info first:
> OS: RHEL6
> Hadoop version: hadoop-1.0.3-1.x86_64
>
> 20 node cluster configured as follows
> 1 node as primary namenode
> 1 node as secondary namenode + job tracker
> 18 nodes as datanode + tasktracker
>
> I have HDFS up and running and have the following in mapred-site.xml:
> <property>
>   <name>mapred.system.dir</name>
>   <value>hdfs://hadoop1/mapred</value>
>   <description>Shared data for JT - this must be in HDFS</description>
> </property>
>
> I have created this directory in HDFS, owner mapred:hadoop,
> permissions 700 which seems to be the most common recommendation
> amongst multiple, often conflicting articles about how to set up
> Hadoop.  Here is the top level of my filesystem:
> hyperion-hdp4@hdfs:hadoop fs -ls /
> Found 3 items
> drwx------   - mapred hadoop          0 2012-10-09 12:58 /mapred
> drwxrwxrwx   - hdfs   hadoop          0 2012-10-09 13:00 /tmp
> drwxr-xr-x   - hdfs   hadoop          0 2012-10-09 12:51 /user
>
> Note, it doesn't seem to really matter what permissions I set on
> /mapred since when the Jobtracker starts up it changes them to 700.
>
> However, when I try to run the hadoop example teragen program as a
> "regular" user I am getting this error:
> hyperion-hdp4@robing:hadoop jar /usr/share/hadoop/hadoop-examples*.jar
> teragen -D dfs.block.size=536870912 10000000000
> /user/robing/terasort-input
> Generating 10000000000 using 2 maps with step of 5000000000
> 12/10/09 16:27:02 INFO mapred.JobClient: Running job:
> job_201210072045_0003
> 12/10/09 16:27:03 INFO mapred.JobClient:  map 0% reduce 0%
> 12/10/09 16:27:03 INFO mapred.JobClient: Job complete:
> job_201210072045_0003
> 12/10/09 16:27:03 INFO mapred.JobClient: Counters: 0
> 12/10/09 16:27:03 INFO mapred.JobClient: Job Failed: Job
> initialization failed:
> org.apache.hadoop.security.AccessControlException:
> org.apache.hadoop.security.AccessControlException: Permission denied:
> user=robing, access=EXECUTE, inode="mapred":mapred:hadoop:rwx------
> at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
> at
> sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:39)
> at
> sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:27)
> at java.lang.reflect.Constructor.newInstance(Constructor.java:513)
> at
> org.apache.hadoop.ipc.RemoteException.instantiateException(RemoteException.java:95)
> at
> org.apache.hadoop.ipc.RemoteException.unwrapRemoteException(RemoteException.java:57)
> at
> org.apache.hadoop.hdfs.DFSClient$DFSOutputStream.<init>(DFSClient.java:3251)
> at org.apache.hadoop.hdfs.DFSClient.create(DFSClient.java:713)
> at
> org.apache.hadoop.hdfs.DistributedFileSystem.create(DistributedFileSystem.java:182)
> at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:555)
> at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:536)
> at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:443)
> at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:435)
> at
> org.apache.hadoop.security.Credentials.writeTokenStorageFile(Credentials.java:169)
> at
> org.apache.hadoop.mapred.JobInProgress.generateAndStoreTokens(JobInProgress.java:3537)
> at
> org.apache.hadoop.mapred.JobInProgress.initTasks(JobInProgress.java:696)
> at org.apache.hadoop.mapred.JobTracker.initJob(JobTracker.java:4207)
> at
> org.apache.hadoop.mapred.FairScheduler$JobInitializer$InitJob.run(FairScheduler.java:291)
Why are you using a HDFS-based directory for mapred.system.dir?
If you want to share this directory, a good approach to this, is to use
NFS for
that.

Marcos Luis Ort�z Valmaseda
*Data Engineer && Sr. System Administrator at UCI*
about.me/marcosortiz <http://about.me/marcosortiz>
My Blog <http://marcosluis2186.posterous.com>
Tumblr's blog <http://marcosortiz.tumblr.com/>
@marcosluis2186 <http://twitter.com/marcosluis2186>

10mo. ANIVERSARIO DE LA CREACION DE LA UNIVERSIDAD DE LAS CIENCIAS INFORMATICAS...
CONECTADOS AL FUTURO, CONECTADOS A LA REVOLUCION

http://www.uci.cu
http://www.facebook.com/universidad.uci
http://www.flickr.com/photos/universidad_uci