I just tested out Accumulo 1.4.2 with HDP 1.1. It appears to be working. The only issue I've encountered is that the monitor overview page is somewhat underpopulated as it doesn't have permission to pull in information about HDFS and the NameNode. Here is a rough outline of the steps I followed:
- installed HDP according to the instructions at http://hortonworks.com/hdp11-hmc-quick-start-guide/ (deselected all optional services, otherwise kept defaults) - created accumulo user with group accumulo on all nodes and set up passwordless ssh - used the hdfs user to create a folder that accumulo is allowed to write to (can be accomplished with sudo /usr/lib/hadoop/sbin/hadoop-create-user.sh -u accumulo) - installed zookeeper manually - installed accumulo manually * copied an example configuration and customized accumulo-env.sh, accumulo-site.xml, masters, slaves, etc. as usual * set instance.dfs.dir to /user/accumulo/accumulo in accumulo-site.xml Billie
