A couple of months ago our friends at Cfengine presented a brief overview of the possibilities of a Cfengine-managed OpenNebula setup at the Large Installation System Administration (LISA) conference in San Jose. The Cfengine team presented how Cfengine may be used on both the physical and virtual sides of an OpenNebula-based cloud. More specifically, they presented how Cfengine can be used to install and configure the physical infrastructure in an OpenNebula cloud, followed by the launch and configuration of generic virtual machine images that will run on top of that OpenNebula infrastructure.

In a recent news article, Cfengine has announced that its Orion Cloud Pack, which was originally conceived to make the use of the Amazon EC2 Cloud simple for users of Cfengine, is also working on OpenNebula cloud instances.

In July 2010 the StratusLab project conducted two surveys to collect requirements for the StratusLab cloud distribution and to understand the existing experience with virtualization/cloud technologies in Europe. Funded through the European Union Seventh Framework Programme (FP7), StratusLab is a two year project aimed to successfully integrate ‘cloud computing’ technologies into ‘grid’ infrastructures.
The survey results are presented in project deliverable D2.1 Review of the Use of Cloud and Virtualization Technologies in Grid Infrastructures. The project conducted two online surveys, one aimed at system administrators and the other at users.  Over two-thirds of sysadmins, and over three-quarters of users surveyed intend to use cloud.  Perhaps more surprisingly, over one third of both groups are already using cloud technologies right now. The most popular public clouds are Amazon Web Services and Google App Engine, while OpenNebula is the most popular open source tool for cloud computing management.
The StratusLab surveys have identified certain trends and requirements for cloud technologies among the Grid community that will be addressed to provide the first full cloud solution for grid and cluster computing.  The StratusLab project aims to produce a toolkit to cloud-enable Grid infrastructures based on OpenNebula. The first version of its cloud computing distribution was released few weeks ago.

I am happy to announce first release of the Python OCA bindings. These bindings wrap OpenNebula’s XML-RPC methods in the Python objects, which allows developers to interact with OpenNebula in a more pythonic way.

The package is available on pypi so if you want to try it just run:

$ easy_install oca

Or download the code from github and install it by running:

$ python setup.py install

Here is an example that shows how you can add new host using Python bindings:

[python]
#!/usr/bin/env python

client = oca.Client(‘user:password’, ‘http:12.12.12.12:2633/RPC2’)
new_host_id = oca.Host.allocate(client, ‘host_name’, ‘im_xen’, ‘vmm_xen’, ‘tm_nfs’)
hostpool = oca.HostPool(client)
hostpool.info()
for i in hostpool:
if i.id == new_host_id:
host = i
break
print host.name, host.str_state
[/python]

For more details how to use Python OCA read the documentation

Try it and share your thoughts, any feedback is welcome.

C12G Labs announced today a major new release of OpenNebulaPro, the enterprise-ready edition of the OpenNebula Toolkit. OpenNebula 2.0 is the most flexible and innovative enterprise-class cloud computing tool for managing a data center’s virtual infrastructure. This fully open-source toolkit includes production-ready features for integration, management, scalability, security and accounting that many enterprise IT shops need for private and hybrid cloud adoption. OpenNebulaPro includes the most recent thoroughly tested and quality-controlled version of OpenNebula and tested software extensions from the community (ecosystem and addons catalogs) for its operation in business environments.

OpenNebulaPro is provided under open-source license to customers and partners on an annual subscription basis through the new OpenNebula.pro Support Portal. The subscription model brings several additional benefits in terms of long term support, production-level support with professional SLAs, integration support for optimal and scalable execution in any hardware and software combination, certification support to validate compatibility with complementary components and customizations, regular updates and upgrades, and additional tools for the administration of OpenNebula cloud instances in production environments.

C12G has created a new howto to explain the way to use OpenNebula with qcow images. Using them has the benefit of occupying less space, faster cloning time and solving problems related with sparse images. It is also a nice example on how OpenNebula behavior can be changed to suit the system administrator or infrastructure needs, in these case the storage model. There are still unexplored qcow feature besides the ones described in the text but it serves as the basis to implement them.

OpenNebula has been designed to be easily adapted to any infrastructure and easily extended with new components.  The result is a modular system that can implement a variety of Cloud architectures and can interface with multiple datacenter services. In the new white paper OpenNebula APIs, C12G reviews the main interfaces of OpenNebula, their use and gives pointers to additional documentation for each one. Interfaces are classified in two categories: end-user cloud and system interfaces. Cloud interfaces are primary used to develop tools targeted to the end-user, and they provide a high level abstraction of the functionality provided by the Cloud. On the other hand, the system interfaces expose the full functionality of OpenNebula and are mainly used to adapt and tune the behavior of OpenNebula to the target infrastructure.

The Management of Data Information, and Knowledge Group (MaDgIK) at the University of Athens focuses on several research areas, such as Database and Information Systems, Distributed Systems, Query Optimization, and Digital Libraries. During the past few years, projects within this group started offering and sharing hardware resources through a virtualized infrastructure. We eventually built our own IaaS-cloud using open source software, namely Xen, Debian and OpenNebula. However, that was not enough, since we needed custom solutions to suit our needs. At that point we called upon Eolus, the god of the winds, to blow and shape the clouds.

Eolus is our open source attempt to join the forces of OpenNebula and Java Enterprise Edition. It is far from being an end product, yet it provides functionality that serves our purposes. In short, OpenNebula is used as a management tool for virtual resources that we exploit in building higher level custom services available through a JEE application container. An advanced VM scheduler called Nefeli and a web based administration console are only a couple of such high level components we offer to users. Our success stories include undergraduate theses, researchers and European funded projects (e.g. D4Science) experimenting and exploiting cloud resources. You can have a glimpse of our efforts, released under the EUPL licence, at https://code-projects.di.uoa.gr/eolus.

Here at OpenNebula we’re really happy with the efforts of our community contributor Damien Raude-Morvan, who is responsible for the new ‘opennebula’ package in the Debian Sid release. This package features the recent OpenNebula 2.0.1 plus some cool features contributed by Damien:

  • OpenNebula Java API package.
  • A modified oned.conf to work out of the box with Debian.
  • A tuned up onehost command which displays useful information for new OpenNebula users.
  • A README file with an example session with OpenNebula, very useful for new users.
  • Tested following the Debian policy. Thanks to Łukasz Oleś for testing it under Xen!

We have a strong commitment with this initiative and we plan to offer our support to keep this package up to date, especially with the upcoming OpenNebula releases.

You can follow this packaging effort at the Alioth OpenNebula Project’s page, you can also clone the git repo or subscribe to the mailing list.

So go ahead and try OpenNebula on Debian Sid! It’s now as easy as running:

apt-get install opennebula

3rd December, 2010. A month after the release of OpenNebula 2.0 the OpenNebula team is pleased to announce its first maintenance release. This update is the first in a series of stabilization updates to 2.0.0, coming regularly. OpenNebula 2.0.1 brings bugfixes and it is a recommended update for everyone running OpenNebula 2.0 or earlier versions. As the release only contains bugfixes, it will be a safe and smooth update for 2.0 users. OpenNebula 1.4 users may be interested in the following scripts to migrate to 2.0.1, contributed by Marlon Nerling.

Some OpenNebula 2.0.1 fixes include:

  • Better packaging of OpenNebula, including its integration in Debian unstable and improved rpm’s.
  • Improved distribution of remote actions
  • Support for drivers in the DISK attribute to use qcow formats or custom Xen drivers
  • Several minor bug fixes. Check the OpenNebula 2.0.1 issues for a complete list.

Also you may be interested in the full list of changes introduced with OpenNebula 2.0, more details here.

As usual for OpenNebula releases, this release is name after a nebula, in this case Abell 39 which is an spherical nebula about five light-years across, discovered by George Abell in 1966.

C12G Labs has just announced the launch of the new OpenNebula.pro Support Portal, which centralizes all support resources for OpenNebulaPro, the commercially supported distribution of the OpenNebula toolkit. The OpenNebula.pro support portal provides the professional support services and tools to integrate, build, certificate and manage enterprise-ready cloud infrastructures and solutions.
OpenNebula is a fully open-source technology. OpenNebulaPro is a commercially supported distribution of OpenNebula with selected stable and tested software extensions and the patches available for its operation in business environments. OpenNebulaPro represents the best balance of maturity and features selected based on their appropriateness for commercial deployment.