This repository contains the set of scripts used to setup a Spark cluster on your own Openstack installation. These scripts are intended to be used by special image prepared for Openstack and is not expected to work on other AMIs. This special image will be available soon when we finish testing. Instructions for use are in progress.
The Spark cluster setup is guided by the values set in ec2-variables.sh
.setup.sh
first performs basic operations like enabling ssh across machines, mounting ephemeral
drives and also creates files named /root/spark-openstack/masters
, and
/root/spark-openstack/slaves
.
Following that every module listed in MODULES
is initialized.
To add a new module, you will need to do the following:
a. Create a directory with the module's name
b. Optionally add a file named init.sh
. This is called before templates are configured
and can be used to install any pre-requisites.
c. Add any files that need to be configured based on the cluster setup to templates/
.
The path of the file determines where the configured file will be copied to. Right now
the set of variables that can be used in a template are
{{master_list}}
{{active_master}}
{{slave_list}}
{{zoo_list}}
{{cluster_url}}
{{hdfs_data_dirs}}
{{mapred_local_dirs}}
{{spark_local_dirs}}
{{default_spark_mem}}
You can add new variables by modifying deploy_templates.py
d. Add a file named setup.sh
to launch any services on the master/slaves. This is called
after the templates have been configured. You can use the environment variables $SLAVES
to
get a list of slave hostnames and /root/spark-openstack/copy-dir
to sync a
directory across machines.
e. Modify https://github.com/mesos/spark/blob/master/openstack/spark_openstack.py to add your module to the list of enabled modules (it will be valid after spark team will approve our code).