Be cautious. The assumption with all techniques using the Ambari REST API to auto start the cluster is the notion that the ambari-server process is running and also reachable via the network. I have seen cases in AWS where the network connection is not yet available to the Ambari server host, resulting in the REST calls failing. ... View more

Is Kerberos also in play? The docs page here gives an example for connecting with the HBase service principal. Have you tried a connection URL like? jdbc:phoenix:<Zookeeper_host_name>:<port_number>:<secured_Zookeeper_node>:<user_name> ... View more

Due to the wide variety of drive configurations utilized in DataNodes, the failure tolerance for disks is configurable. The dfs.datanode.failed.volumes.tolerated property in hdfs-site.xml allows you to specify how many volumes are lost before the DataNode is marked offline. Once the node is offline, the Namenode will use information in the block reports to create new replicas for any blocks left in an under-replicated state by the failure. You can tune the HDFS DataNode storage alert warning setting to account for the minimum amount of storage you desire to have available. This will give you a warning before things go critical. Individual disk monitoring is usually handled by an enterprise-level monitoring tool such as OpenView, etc. ... View more

@Madhan Neethiraj this is an HDP 2.2.8 install and creating a username that starts with an alphabetic character does work. ... View more

It was not a length problem. The user is not a member of the group he is being added into. There is an odd mix of LDAP users and local groups in the test environment. You should enter the length as the answer since that is the question I asked. Then I can accept the answer. Thanks. ... View more

The response from the UI is "enter a valid user name". I will try again. ... View more

No. Not my question, but I have done remote repo setups without EPEL and just HDP and HPD-UTILS with no problems. ... View more