mikewalch / amara

Starts and manages Accumulo in Marathon

Amara

Starts and manages an Accumulo instance in Marathon

Installation

Clone Amara on cluster where you have HDFS and Zookeeper running.

git clone https://github.com/mikewalch/amara.git

Configuration

1. Create a config file and edit it for your environment. If you have JAVA_HOME, HADOOP_PREFIX, and ZOOKEEPER_HOME set as environment variables in your shell, these values do not need to be set in config.json as Amara will use what is set in your environment.

    cd amara/conf
    cp config.json.example config.json
    vim config.json
   

2. Download an Accumulo release tarball that matches the AccumuloVersion set in config.json and copy it to the upload/ directory.

    cp accumulo-1.7.1-bin.tar.gz /path/to/amara/upload
   

Start and manage Accumulo

Initialize Accumulo

cd amara/
./bin/amara init

Start Accumulo in Marathon

./bin/amara start

Destroy Accumulo in Marathon

./bin/amara destroy