Error in Hbase backup from secure to normal cluster.

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Error in Hbase backup from secure to normal cluster.

mathes waran
Hi,

I tried distcp method in hbase backup from secure to normal cluster,In my backup job the data is transferred successfully from secure to normal cluster,but in resource manager job id is shown as failed due to error of Non super user cannot change owner.

However, table and snapshot is created successfully and restoring table from normal to secure is executed successfully.

But in HBase backup due to this error ,application Id is shown as failed.

Please find the error details as below:

Caused by: org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.security.AccessControlException): Non-super user cannot change owner

Even I entered super user credentials while executing distcp job from secure to normal cluster,this exception is occurred.

please note that,this error is occurred only on backup from secure to normal,and its not occur on restoring table.

Can u pls tell me is there any changes in configuration file or how could i solve this..?

Looking back.!

Thanks,

Matheskrishna

Reply | Threaded
Open this post in threaded view
|

Re: Error in Hbase backup from secure to normal cluster.

Rakesh Radhakrishnan-2
Hi,

Hope you are executing 'distcp' command from the secured cluster. Are you executing the command from a non-super user? Please explain me the command/way you are executing to understand, how you are entering "entered super user credentials" and -D command line args. 

Also, please share your hdfs-site.xml, core-site.xml configurations.

Have you modified any of these configurations or passed as -D command line args.

In core-site.xml on the secure cluster side:
<property>
  <name>ipc.client.fallback-to-simple-auth-allowed</name>
</property>

In hdfs-site.xml on the secure cluster side:
<property>
  <name>dfs.permissions.superusergroup</name>
  <description>This is by default set to an arbitrary string "superuser", which is
    mostly a non existing group on most (all?) environments. Changing this
    and/or creating such a group name on the NN machine will let you
    permit more users to act as superusers, if needed.
 </description>
</property>

Again, I'd suggest you to post this problem to the hbase user mailing list to know any specific configurations from HBase side.

Regards,
Rakesh

On Mon, Jul 11, 2016 at 3:45 PM, mathes waran <[hidden email]> wrote:
Hi,

I tried distcp method in hbase backup from secure to normal cluster,In my backup job the data is transferred successfully from secure to normal cluster,but in resource manager job id is shown as failed due to error of Non super user cannot change owner.

However, table and snapshot is created successfully and restoring table from normal to secure is executed successfully.

But in HBase backup due to this error ,application Id is shown as failed.

Please find the error details as below:

Caused by: org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.security.AccessControlException): Non-super user cannot change owner

Even I entered super user credentials while executing distcp job from secure to normal cluster,this exception is occurred.

please note that,this error is occurred only on backup from secure to normal,and its not occur on restoring table.

Can u pls tell me is there any changes in configuration file or how could i solve this..?

Looking back.!

Thanks,

Matheskrishna


Reply | Threaded
Open this post in threaded view
|

Re: Error in Hbase backup from secure to normal cluster.

Rakesh Radhakrishnan-2
Hi Matheskrishna,

Adding one more thought to my above comments.

Since you are telling about the job execution failure on distcp, I think it would be good to analyse the failure logs of the job to see any problem is that inside the staging directory in hdfs or we may get some hint about the cause.

Regards,
Rakesh

On Mon, Jul 11, 2016 at 6:13 PM, Rakesh Radhakrishnan <[hidden email]> wrote:
Hi,

Hope you are executing 'distcp' command from the secured cluster. Are you executing the command from a non-super user? Please explain me the command/way you are executing to understand, how you are entering "entered super user credentials" and -D command line args. 

Also, please share your hdfs-site.xml, core-site.xml configurations.

Have you modified any of these configurations or passed as -D command line args.

In core-site.xml on the secure cluster side:
<property>
  <name>ipc.client.fallback-to-simple-auth-allowed</name>
</property>

In hdfs-site.xml on the secure cluster side:
<property>
  <name>dfs.permissions.superusergroup</name>
  <description>This is by default set to an arbitrary string "superuser", which is
    mostly a non existing group on most (all?) environments. Changing this
    and/or creating such a group name on the NN machine will let you
    permit more users to act as superusers, if needed.
 </description>
</property>

Again, I'd suggest you to post this problem to the hbase user mailing list to know any specific configurations from HBase side.

Regards,
Rakesh

On Mon, Jul 11, 2016 at 3:45 PM, mathes waran <[hidden email]> wrote:
Hi,

I tried distcp method in hbase backup from secure to normal cluster,In my backup job the data is transferred successfully from secure to normal cluster,but in resource manager job id is shown as failed due to error of Non super user cannot change owner.

However, table and snapshot is created successfully and restoring table from normal to secure is executed successfully.

But in HBase backup due to this error ,application Id is shown as failed.

Please find the error details as below:

Caused by: org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.security.AccessControlException): Non-super user cannot change owner

Even I entered super user credentials while executing distcp job from secure to normal cluster,this exception is occurred.

please note that,this error is occurred only on backup from secure to normal,and its not occur on restoring table.

Can u pls tell me is there any changes in configuration file or how could i solve this..?

Looking back.!

Thanks,

Matheskrishna



Reply | Threaded
Open this post in threaded view
|

Re: Error in Hbase backup from secure to normal cluster.

Rakesh Radhakrishnan-2
>>>->Now I executed the hbase backup command with super user credentials ,job are submitted in resource manager,while executing it throws an error of "Non super user cannot change owner" ,at same time backup is completed,table is stored in destination cluster along with error status.(Job Id is also faied in resource manager)

Could you please share the job execution logs, probably will get some hints about the failure. Also, please share the complete command structure you are executing for hbase backup.

Thanks,
Rakesh

On Tue, Jul 12, 2016 at 11:42 AM, mathes waran <[hidden email]> wrote:
Hi Rakesh Radhakrishnan,

       yes,I am executed this command from only secure cluster ,and I executed job with super user credentials.
please find the details below which I followed:
->set ipc.client.fallback-to-simple-auth-allowed as true in HBase-site.xml for the purpose of allowing of data from secure to normal cluster.
->set property hdfs-site.xml on the secure cluster side:

 <property>
 <name>dfs.permissions.superusergroup</name>
 <value>supergroupname</value>
 </property>

->Now I executed the hbase backup command with super user credentials ,job are submitted in resource manager,while executing it throws an error of "Non super user cannot change owner" ,at same time backup is completed,table is stored in destination cluster along with error status.(Job Id is also faied in resource manager)

->For your information,while executing restore command the table is restored successfully without this kind of issues.

So please could you explain how to solve this exception.

Looking back for your response,

Thanks,
Matheskrishna


On Mon, Jul 11, 2016 at 6:32 PM, Rakesh Radhakrishnan <[hidden email]> wrote:
Hi Matheskrishna,

Adding one more thought to my above comments.

Since you are telling about the job execution failure on distcp, I think it would be good to analyse the failure logs of the job to see any problem is that inside the staging directory in hdfs or we may get some hint about the cause.

Regards,
Rakesh

On Mon, Jul 11, 2016 at 6:13 PM, Rakesh Radhakrishnan <[hidden email]> wrote:
Hi,

Hope you are executing 'distcp' command from the secured cluster. Are you executing the command from a non-super user? Please explain me the command/way you are executing to understand, how you are entering "entered super user credentials" and -D command line args. 

Also, please share your hdfs-site.xml, core-site.xml configurations.

Have you modified any of these configurations or passed as -D command line args.

In core-site.xml on the secure cluster side:
<property>
  <name>ipc.client.fallback-to-simple-auth-allowed</name>
</property>

In hdfs-site.xml on the secure cluster side:
<property>
  <name>dfs.permissions.superusergroup</name>
  <description>This is by default set to an arbitrary string "superuser", which is
    mostly a non existing group on most (all?) environments. Changing this
    and/or creating such a group name on the NN machine will let you
    permit more users to act as superusers, if needed.
 </description>
</property>

Again, I'd suggest you to post this problem to the hbase user mailing list to know any specific configurations from HBase side.

Regards,
Rakesh

On Mon, Jul 11, 2016 at 3:45 PM, mathes waran <[hidden email]> wrote:
Hi,

I tried distcp method in hbase backup from secure to normal cluster,In my backup job the data is transferred successfully from secure to normal cluster,but in resource manager job id is shown as failed due to error of Non super user cannot change owner.

However, table and snapshot is created successfully and restoring table from normal to secure is executed successfully.

But in HBase backup due to this error ,application Id is shown as failed.

Please find the error details as below:

Caused by: org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.security.AccessControlException): Non-super user cannot change owner

Even I entered super user credentials while executing distcp job from secure to normal cluster,this exception is occurred.

please note that,this error is occurred only on backup from secure to normal,and its not occur on restoring table.

Can u pls tell me is there any changes in configuration file or how could i solve this..?

Looking back.!

Thanks,

Matheskrishna