geraldpereira / api-examples Goto Github PK

#Using GreenQloud Public APIs

###GreenQloud: Compute and Storage on demand

GreenQloud offers a compute API that can be used to create and run virtual servers to run software which is called ComputeQloud. GreenQloud also offers an object/web storage service that can be used to store arbitrary files or data which is called StorageQloud.

###ComputeQloud

ComputeQloud is GreenQloud's compute service with an API compatible with Amazon's EC2 service. There are several command line tools that can be used to call the service. The most popular open source tool is called euca2ools, but GreenQloud also supports Amazon's official ec2-api-tools.

###StorageQloud

StorageQloud is GreenQloud's storage service with an API compatible with Amazon's S3 service. There are several command line tools that can be used to call the service. Examples of these are s3cmd and s3sync.

###Installation of client tools

###Euca2ools

Instructions to install euca2ools for GreenQloud can be found here

A User Guide of euca2ools can also be found here

Once the tool is installed one can run commands such as:

euca-describe-images

To list up available OS images

And to run up a virtual machine one can run a command such as:

euca-run-instances -k default -t t1.micro qmi-9ac92558

to start up an Ubuntu 11.04 instance of size t1.micro

To get the address of this instance type:

euca-describe-instances

INSTANCE	i-xxxxx	qmi-9ac92558	i-46-149-xxx-xxx.compute.is-1.greenqloud.com		running 	default 	0 	 	t1.micro 	2012-05-23T16:53:50.000+00:00 	is-1a

And to log into this instance with ssh you have to pass in the private key used to start up the instance (in the example above the key 'default' was used):

ssh -i [path_to_default_private_key] [email protected]

###ec2-api-tools

Instructions for the Amazon tools can also be found here. GreenQloud supports most commands of the API, and a list of supported commands can be found here.

Similarly as with the euca2ools the ec2 api command to start up an instance can be done with:

ec2-run-instances -k default -t t1.micro qmi-9ac92558

###S3cmd

Instructions to install s3cmd for Ubuntu can be found here and for CentOs here

Once the tool is installed one can run command such as:

s3cmd ls

to list up buckets for the account.

To create a storage bucket with name testbucket type:

s3cmd mb s3://testbucket

And to remove this bucket call:

s3cmd rb s3://testbucket

To upload a file to the file 'testfile' to bucket 'testbucket' call:

s3cmd put testfile s3://testbucket

For other command get the help by:

s3cmd --help

Note that bucket names are unique globally, so no user can have the same bucket name as another.

###S3Sync

S3sync is a solution to sync folders towards an S3 compatible storage service such as StorageQloud

Instructions to set up for Linux can be found on:

http://support.greenqloud.com/entries/20210281-using-s3sync-with-the-storage-api

Instructions to set up for Windows/.Net can be found on

http://sprightlysoft.com/blog/?p=94

###Other Great Tools

There are other tools that may be used to call GreenQlouds APIs.

Awssum is a Node.js module that specifically supports GreenQloud's Compute and Storage services. GreenQloud's own management console, http://my.greenqloud.com, is built on Awssum and the revolutionary Meteor framework. Using the two together with a Meteorite smart package is a very fast way to build a web application:

Awssum - examples: http://awssum.io
Meteor: http://meteor.com
Smart package: https://github.com/possibilities/meteor-awssum

Here's how to get started on OSX with Awssum, Node.js, Meteor and Meteorite.

Prerequisite: If you don't have command line build tools like gcc installed you need to start with installing XCode and the command line tools:

1. Install XCode from the OSX App Store
2. Open Xcode and go to the Preferences and the Downloads tab
Install the Command Line Tools (cli)
3. Tell OSX to use the latest XCode
sudo xcode-select -switch /Applications/Xcode.app/Contents/Developer
4. To install Node.js it is easiest to do with Homebrew, http://mxcl.github.com/homebrew/.
ruby -e "$(curl -fsSkL raw.github.com/mxcl/homebrew/go)"

Now install the frameworks (from your home directory):

# install node, npm, meteor & meteorite & awssum
brew install node
curl https://npmjs.org/install.sh | sh
curl https://install.meteor.com | sh
sudo npm install -g meteorite
npm install awssum

Finally create your app and add the awssum smart package:

mrt create myGQApp
cd myGQApp
mrt update
mrt add awssum
mrt (runs the app on http://localhost:3000)

To use the Compute Qloud and Storage Qloud APIs you have to initialize them on the server side e.g. in the default server side code block of myGQApp.js :

if (Meteor.isServer) {...

var greenqloud = AWSSum.load('greenqloud/greenqloud');
var SQ = AWSSum.load('greenqloud/s3','S3');
var CQ = AWSSum.load('greenqloud/ec2', 'Ec2');

function getSQ(apiKey, secretKey, region) {
	if (!region) {
		region = greenqloud.IS_1;
	}

	return new SQ({
		accessKeyId     : apiKey,
		secretAccessKey : secretKey,
		region			: region
	});
}

function getCQ(apiKey, secretKey, region) {
	if (!region) {
		region = greenqloud.IS_1;
	}

	return new CQ({
		accessKeyId     : apiKey,
		secretAccessKey : secretKey,
		region			: region
	});
}

function encodeUTF8( s ){
	return unescape( encodeURIComponent( s ) );
}
function decodeUTF8( s ){
	return decodeURIComponent( escape( s ) );
}

...

There are many examples on http://awssum.io just remember Compute Qloud is EC2 compatible and Storage Qloud is S3 compatible! Now you can do stuff like:

function getBucketObjects(apiKey, secretKey, bucketName, prefix){
	this.unblock(); //so we can do parallel stuff, meteor is serial by default
	console.log("[StorageQloud] Listing objects in this bucket");


	var bucketOptions = {
		BucketName : bucketName,
		Delimiter : '/',
		Prefix : encodeUTF8(prefix)
	};

	var result = getSQ(apiKey, secretKey).ListObjects(bucketOptions);
	if(result.error){
		return new Meteor.Error(500, result.error);
	}
	else{
		console.log(result.Body.ListBucketResult, 'Result');
		return result.Body.ListBucketResult;
	}
}

Boto is a popular Python library for calling EC2/S3 services:

https://github.com/boto/boto

Sample Code for accessing GreenQloud through python/boto can be found in the folder within this github repository (You will have to modify it by providing your API/Secret keys):

greenqloud-boto-sample
python greenqloud_simple_test.py

Within this github repository

AWS Java SDK is Amazons library for calling EC2/S3 that works with Greenqloud:

http://aws.amazon.com/sdkforjava/

Sample Code for accessing GreenQloud through Java can be found in the folder within this github repository (You will have to modify it by providing your API/Secret keys):

GreenQloud-JavaSDKGettingStarted
ant GettingStartedApp

Cyberduck is a popular GUI (Mac/Windows) tool for accessing StorageQloud:

http://cyberduck.ch/

Instructions for adding a GreenQloud Profile in Cyberduck can be found on http://trac.cyberduck.ch/wiki/help/en/howto/greenqloud

CloudBerry is a popular Windows client to use GreenQloud's StorageQloud:

http://www.cloudberrylab.com/

s3fs is a tool for mounting s3 compatible storage such as StorageQloud as a filesystem directory structure on Linux (through FUSE).

http://code.google.com/p/s3fs/

Instructions can be found in the same articles as describing the installation of s3cmd here

###Notes

Other tips and tricks can be found on GreenQlouds support site:

http://support.greenqloud.com/

For more information about GreenQloud visit:

http://www.greenqloud.com/

The GreenQloud team