Skip to content

CloudScale-Project/Deployment-Scripts

Repository files navigation

CloudScale deployment scripts

CloudScale deployment scripts are Python scripts for deploying showcase for CloudScale project on Amazon Web Services and Openstack. The showcase is a book store written in Java Spring Framework according to TPC-W standard. We call it CloudStore. Scripts are configurable so you can also use them for deploying your application to Amazon Web Services or Openstack.

We also provide the distributed JMeter scripts for load testing the CloudStore application or your own application.

Prerequisites

Using deployment scripts requires that you have already installed the following software:

  • Python 2.7
  • pip
  • virtualenv
  • MySQL database

Installation

  1. Create new virtual environment using Python package virtualenv

  2. In newly created environment execute the following command:

$ pip install -e https://github.com/CloudScale-project/Showcase/deployment-scripts/zipball/deployment-scripts

You can also install it manually by downloading ZIP archive and execute:

$ python setup.py install

from extracted ZIP archive.

This will install CloudScale deployment scripts to your site-packages folder of your Python distribution.

Configuration

Deployment scripts enables user to deploy CloudStore on Amazon Web Services or OpenStack.

We already provide example config files for both clouds and you can find them in bin/ directory. On OpenStack you can also deploy CloudStore using MongoDB (alpha).

Amazon Web Services configuration file

For deploying CloudStore on Amazon Web Services use bin/config.aws.example.ini config file which is structured like:

[DATABASE]

name - Database name

user - Username used to authenticate to database

password - Password used to authenticate to database

dump_url - URL to the database dump. See also

[APPLICATION]

distribution_url - URL to compiled and packaged version of CloudStore.

connection_pool_size - Connection pool size to database. If using multiple databases, value should tell maximum number of connection to one database.

num_instances - Number of database instances

[AUTO_SCALABILITY]

enabled - Possible values: 'yes' or 'no'. Enables or disable autoscalability

cooldown - How many seconds does autoscaling waits before executing new autoscaling operation. See AWS documentation

[AWS]

aws_access_key_id - AWS access key

aws_secret_access_key - AWS secret key

region - Codename of the region where all instances should be deployed

availability_zone - Default availability zone of region where instances should be deployed

[RDS]

instance_type - RDS instance type

num_replicas - Number of RDS replicas

master_identifier - Identifier for master RDS instance

replica_identifier - Identifier for replica RDS instance

[EC2]

instance_type - EC2 instance type

ami_id - Amazon Machine Image from which all EC2 instances will be provisioned

key_name - Name of the key pair on the AWS

key_pair - Path to key pair. It's auto generated

remote_user - Default username to SSH to EC2 instances

instance_identifier - Instance identifier

OpenStack configuration file

For OpenStack we have included support for MySQL database and MongoDB database (alpha).

OpenStack MySQL configuration file structure

[OPENSTACK]

username - Provide username to authenticate to OpenStack

password - Provide password to authenticate to OpenStack

tenant_name - Tenant VMs will launch in

auth_url - Keystone authentication URL

key_name - Name of key pair used to authenticate to VM

key_pair - Path to key pair used to authenticate to VM

[APPLICATION]

num_instances - Number of instances to provision

image_name - Name of the VM image to provision new VMs from

image_username - Name of the user for SSH authentication

instance_type - Flavor used for new instances

[DATABASE]

database_type - Database type (mysql for MySQL and mongo for MongoDB)

database_name - Database name

database_host - Database host

database_user - Database user

database_pass - Database password

instance_type - Flavor used for database instances

num_replicas - Number of replicas without master instance (if replication is used)

connection_pool_size - Connection pool size

[MYSQL]

setup_type - Type of setup (master-slave or master-master)

dump_url - URL to SQL dump

image_name - Name of image used to provision new instances from

showcase_url - URL to the WAR where CloudStore is hosted

OpenStack MongoDB configuration file structure

[OPENSTACK]

Please see description for OpenStack MySQL configuration file structure

[APPLICATION]

Please see description for OpenStack MySQL configuration file structure

[DATABASE]

Please see description for OpenStack MySQL configuration file structure

[MONGODB]

dump_url - URL to MongoDB archive containing exported database

image_name - Name of the image on OpenStack to provision MongoDB instance from

showcase_url - URL to the WAR where CloudStore is hosted

Usage

You can run each script in cloudscale/deployment_scripts/scripts as standalone or use the wrapper around them in bin/run.py. Since these are Python scripts you can run it with python interpreter.

Example for deploying CloudStore on OpenStack:

$ python bin/run.py openstack bin/config.openstack.mysql.example.ini

The general usage for run.py is:

$ python bin/run.py <aws|openstack> <path to config>

Amazon Web Services

Setup RDS service

Before you can run deployment scripts you must setup RDS service. On RDS service create new Parameter Group with name 'mygroup' for MySQL version 5.1 and set the parameter ```query_cache_size`` to value 0.

Run deployment scripts

To deploy showcase on Amazon Web Services edit bin/config.aws.ini file and run:

$ python run.py aws config.aws.ini

Openstack

To deploy MySQL version of showcase on OpenStack edit bin/config.openstack.mysql.ini and run:

$ python run.py openstack config.openstack.mysql.ini

To deploy MongoDB version of showcase on OpenStack edit bin/config.openstack.mongo.ini and run:

$ python run.py openstack config.openstack.mongo.ini

About

CloudScale deployment scripts are Python scripts for deploying showcase for CloudScale project on Amazon Web Services and Openstack.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published