Are you running CentOS 6.6?  There appears to be an issue setting up TLS encryption on this OS:  http://community.cloudera.com/t5/Cloudera-Manager-Installation/Cloudera-Manager-support-for-Centos-6-6/m-p/21478/thread-id/2820/highlight/true   ... View more

Kevin,   1. Log into Cloudera Manager and go to the Hue - Configuration page 2. Search for "hue_safety_valve.ini" to find the Property, Hue Service Advanced Configuration Snippet (Safety Valve) for hue_safety_value.ini 3. Add a snippet similar to the following in the Value textbox:   [librdbms]   [[databases]]     [[[mysqlemployees]]]     nice_name=MySQL - employees     name=employees     engine=mysql     host=<db host>     ## port=     user=<db username>     password=<password>     ##options={}     [[[mysqlworld]]]     nice_name=MySQL - world     name=world     engine=mysql     host=<db host>     ## port=     user=<db username>     password=<password>     ##options={}   4. Save and restart the Hue service   I found the following link that describes the configuration above:  http://gethue.com/dbquery-app-mysql-postgresql-oracle-and-sqlite-query/   "nice-name" is what will be shown in the Hue dropdown, while "name" is the database name.  I have only configured for MySQL, but you can also use engine=postgresql   Chad ... View more

Please refer to the following page:  http://www.cloudera.com/content/cloudera/en/documentation/core/latest/topics/cm_sg_s7_prepare_cluster.html   You either need to prepare your cluster with a user account with a user id greater than min.user.id or change this setting to be less than the default setting of 1000.  Also, the default setting for the banned.users property does not allow the hdfs user to submit and execute jobs.   For CentOS, you can use the following to specify the UID as something over 1000:  useradd -p <passwd> -u 1001 <username> ... View more

Ram,   I also had this issue.  I believe Datameer is not pulling the correct value for yarn.app.mapreduce.am.staging-dir, as my staging jobs were being created under hdfs:/tmp/hadoop-yarn/staging/datameer, rather than /user/datameer.   Try adding the following to your Datameer Hadoop Cluster settings under Custom Properties:   yarn.app.mapreduce.am.staging-dir=/user   Another good clue I received from Datameer support was to make sure YARN Application Classpath matches the value from 'hadoop classpath' on your cluster.  Mine is:   $HADOOP_CONF_DIR,$HADOOP_COMMON_HOME/lib/*,$HADOOP_COMMON_HOME/*,$HADOOP_HDFS_HOME/lib/*,$HADOOP_HDFS_HOME/*,$HADOOP_YARN_HOME/lib/*,$HADOOP_YARN_HOME/*,$HADOOP_MAPRED_HOME/lib/*,$HADOOP_MAPRED_HOME/*   Good luck,   Chad ... View more