Hi @Cassandra Spencer, I did this recently, and it was easier than expected: Dump the Ambari database (ambari) into an sql file, and use sed or your favorite editor to change all hostnames in the sql file. Then, run "ambari-server reset", drop ambari DB, create it again, and import the edited sql file into Postgres. Of course take all the baskups beforehand. ambarirca database is not affected. Hive DB has no references to hosts. Oozie DB has refs to old hosts but you can leave them since it's about old jobs. Hue DB, if you use Hue has refs to old hosts, and has to be updated like ambari DB. After restarting Ambari with new hostnames I had no issues. If you can keep old hostnames as aliases there is no need to change ambari agents. Otherwise you need to change ambari agent properties, to redirect agents to the new Ambari server. Re iptables, that's unrelated to the above. You have to ensure that old required ports are opened. Starting with Ambari server and Ambari agent ports, and including all others. Also keep all ephemeral ports opened because ZooKeeper and some other components are using them. ... View more

You can set permissions to your user folder like this hdfs dfs -chmod 700 /user/user1 and if the owner of this folder is user1, then only he can list the folder hdfs dfs -ls /user/user1 ... works for user1 but doesn't work for other users However, hdfs as the HDFS super-user can also list it. ... View more

Hi @bpreachuk, according to this page the bulk load tool doesn't have such a feature, but for smaller files, up to "tens of megabytes" you can use a single threaded psql.py tool which can interpret the first line as a list of columns by using the "-h in-line" option. Thinking about the bulk MR tool it's indeed hard to implement this because every mapper gets a chunk of the file, and we'd like only 1 mapper to remove the very first line, so it will have to be marked in a special way. More details about commands here. ... View more

Hi @marko, if I remember correctly Spark-1.6 TP for HDP-2.3.4 was announced 1 day after its GA. Therefore, you can expect Spark-2.0 TP with about the same "delay", and its inclusion in one of the coming versions of HDP soon after. Stay tuned! ... View more