I run all process with the same user. And it should be the superuser since it
start the namenode, is there some configuration I need to do to let the
datanode become superuser?
> On 9 Nov 2018, at 3:37 PM, Attila Bukor <abu...@cloudera.com.INVALID
> <mailto:abu...@cloudera.com.INVALID>> wrote:
>
> All HDFS processes (NameNodes, DataNodes, JournalNodes, FailoverControllers)
> to
> run as the same user, e.g. “hdfs”.
>
>
>> On 2018. Nov 9., at 8:35, Harinder Singh <harindersinghbedi...@gmail.com
>> <mailto:harindersinghbedi...@gmail.com>> wrote:
>>
>> I think you need a superuser to start the service. Is datanode a superuser?
>>
>> Regards
>> Harinder
>>
>> On Thu, Nov 8, 2018 at 11:29 PM ZongtianHou <zongtian...@icloud.com.invalid
>> <mailto:zongtian...@icloud.com.invalid>> wrote:
>> Hi, everyone
>> I set up kerberos for the hdfs cluster, but after I start name node, then
>> the datanode, In the namenode log file, it display the following error:
>>
>> 2018-11-09 15:09:38,725 WARN
>> org.apache.hadoop.security.UserGroupInformation: No groups available for
>> user datanode
>> 4870 2018-11-09 15:09:38,725 INFO org.apache.hadoop.ipc.Server: IPC Server
>> handler 0 on 8020, call
>> org.apache.hadoop.hdfs.server.protocol.DatanodeProtocol.versionRequest from
>> 127.0.0.1:56409 <http://127.0.0.1:56409/> Call#132 Retry#0:
>> org.apache.hadoop.security.AccessControlException: Access denied for user
>> datanode. Superuser privilege is required
>>
>> The kerberos auth for name node and data node are both ok, can anyone see
>> the problem here, any hint will be very appreciated.
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: user-unsubscr...@hadoop.apache.org
>> <mailto:user-unsubscr...@hadoop.apache.org>
>> For additional commands, e-mail: user-h...@hadoop.apache.org
>> <mailto:user-h...@hadoop.apache.org>
>>
>
>
>
>
