Prequistes   if you are doing Prod / Test (Provided if you have huge amount of data in that desired datanode that you are going to decommision.    Increase the Datanode network bandwidh.  Maximum number of replication threads on a DataNode and  Hard limit on the number of replication threads on a DataNode     Click on the Host that has Datanode Role  - Perform Decommision and once the decommissioned is finished  that you can monitor in the Namenode Web UI 50070 .      Sometime I would rather leave it as decommisioned state and later if you want to include the instance back as for scalability you follow the above same procedure instead of decommision u will click recommision that it joins back to the cluster .       if you want to delete the instance after decommisioning please follow the manual    http://www.cloudera.com/documentation/manager/5-1-x/Cloudera-Manager-Managing-Clusters/cm5mc_role_instances.html#cmug_topic_5_2__section_xwx_hvt_cn   if you need more information please let me know  ... View more

did you try using the --force which i mentioned ?  ... View more

 I dont know if you are using the quickstart for learning but I would suggest you to install just the CDH (manage it using commandline) and not use the cloudera manager to manage your clusters. or you have another option which I havent tried because i have a 12GB ram .  How about using the installer .bin file PATH A installation of Cloudera manager in your VM. but it would requires internet as you will be downloding parcels and activating it across the clusters.Also since you are going to run the CM in low RAM resouce you might want to tweak  its maximum heap size by modifying -Xmx parameter in SCM init script /etc/default/cloudera-scm-server and down line just save some resouce you may have to shutdown few service which you are not using .  ... View more

@saisvk   Usually when you CREATE TABLE LIKE syntax you dont specify coloumn name , it will be dervied from the source table.    It just the precidence well try this you should be able to overcome this error      Note -  Make sure you have the write / read access to the path you mention in the LOCATION .    Syntax    CREATE [EXTERNAL] TABLE [IF NOT EXISTS] [db_name.]table_name LIKE { [db_name.]table_name | PARQUET 'hdfs_path_of_parquet_file' } [COMMENT 'table_comment'] [STORED AS file_format] [LOCATION 'hdfs_path'] create table t2 like t1 stored as textfile LOCATION '/user/hduser'; Le me know if that works      ... View more

if I understand clearly you need to remove a datanode from one your nodes ?  Are you performing in testing enviroment  or production clusters ?  whats the replication factor in your cluster Please let me know .  ... View more

Step 1    Open the terminal fire the below command    sudo /home/cloudera/cloudera-manager --express -- Terminal Step 2    Wait patiently 🙂 it will take some time because it has bring down certain service and again come up  you should be able to see this to confirm that all the service are up and runining    Success - Launch Screen    URL - to launch the Cloudera Manager Web UI in the browser   http://quickstart.cloudera:7180     Step 3     after you click agree -> you will prompted in the login screen .  USERNAME - admin  PASSWORD - admin Cloudera Manger WEB UI     Intially all your service / instance will be in stop status You need to start it manually    Please let me know if you need more information.      ... View more

bear with me but did you change the /etc/sysconfig/network file in all the nodes 1 2 3 4 as mentioned earlier ?  did you restart the network  in your os ?    ... View more

@darkdante  were you able to receive heartbeat from agent.   ... View more

see if you have multiple supervisord runining  ps aux | grep supervisor share the logs  to see whats it is complain about supervisord    ... View more

@darkdante   The commuity had some kernel version compatability issue a month back if I remember  You should be  good with 5.12 we are using 5.12 guess what HUE interface has some new functionalties and many more  ... View more