Profile cover photo
Profile photo
Scalable, Modular, Adaptable, Systems Management
Scalable, Modular, Adaptable, Systems Management


Heya everyone,

Have you checked out Singularity yet? While it is not a component of Warewulf, it is a very cool Container solution focused on creating portable application stacks (and written by me, Greg; the author of Warewulf).

It works by collecting an application (or set of applications in a defined workflow) including all runtime dependencies, files, data, and scripts and packaging them all into a executable container. The application (or workflow) within the container can then be run by either installing them into a user's Singularity environment or by executing the container directly. All arguments, and IO (including stdin) will be direct to the application(s) within the container, and these applications can also directly access the files outside of the container (unless you invoke it with the --contain option).

Check out Singularity here:

Have a great holiday!
Add a comment...

Post has attachment
This is a Discussion focused on the Warewulf High Performance Computing platform. Invitees to the HOA are developers, users and stakeholders.

If you would like to be part of this discussion or you have any topics or questions that you would like to have discussed please comment on this hangout event!

Warewulf is a scalable systems management suite originally developed to manage large high-performance Linux clusters. Focused on general scalable systems management, it includes a framework for system configuration, management, provisioning/installation, monitoring, event notification, and more via a modular plugin architecture. Install the components and features you need or leverage the existing system configurations stored within Warewulf to create custom solutions to meet your particular needs.

Warewulf is a flexible solution that has proven itself to be scalable and easy to use.
Add a comment...

Post has attachment
This is a hangout focused on hearing from some of the stakeholders, developers, and active users of the Warewulf cluster management community.

I am starting out with proposing a basic standing agenda as follows:

1. Site updates and use cases (as wanted)
2. Discussion of any issues currently affecting Warewulf users
3. Current status of development, contributions and Q&A
4. Open discussion about future direction, wants, needs, and necessity

Additional discussion points for this meeting would include provisioning models ("of the future"), the architecture platform and contributions of development and documentation.

Please comment on this event to include your thoughts on what topics should be included in this meeting.

Don't forget to RSVP and see you soon!

Add a comment...

Would anyone be interested in a monthly Google Hangout (on Air??) to discuss updates, discussions, feature requests, help, or anything else that might be interesting to users of Warewulf?

Hangouts on Air would be a really great platform for also hosting interactive sessions and presentations as well!
Add a comment...

Post has shared content
Originally shared by ****
Just finished my talk on +Warewulf  #NHC  for #MoabCon2014 .  I still need to wrap up the documentation updates and get the release out the door, but the talk went very well.  So grateful to +Adaptive Computing for allowing me to present again this year and always being so supportive of me and of the project.  I really appreciate all the opportunities they've given me year after year to engage with their customers and really build a solid community of users.  NHC wouldn't be anything close to what it is today without them, so THANK YOU ADAPTIVE!  (Special shoutouts to +Jill King +Ian Nate +Ken Nielson and +Rick McKay!)

And thanks to everyone who attended and to all the NHC users out there who have made it as successful as it is!

PS:  The talk was recorded, so as always, I'll post the video as soon as it's made available.  :-)
Add a comment...

Post has attachment

Post has attachment
Warewulf package release announcement! All releases are available in both source form as well as binary packages for Red Hat compatibles.

Below is an brief overview of release updates:

   * Major backend API architecture updates
   * Refactoring of configuration file parser
   * Many Database scalability optimizations
   * Major bug fixes and updates
   * UI fixes and updates

   * Updates for new API

   * Updates for new API
   * Support for remote node management via submasters
   * Caching optimizations for VNFS and files
   * File provisioning fixes and optimizations
   * Inclusion of ext3 and ext4 for stateful provisioning

   * New architecture for chroot generation using a modular framework

Please goto for more information and downloads!
Add a comment...

Post has attachment
In the first article by Jeff Layton he described how to setup a base Warewulf cluster provisioning system. In this one Jeff continues on speaking about NFS, hybridization, user management, helpful tools and NTP.
Add a comment...

Post has attachment

Post has attachment
Everywhere humans go, they create communities out of diverse, and sometimes hostile, populations. It is a great gift and a terrible that cannot be abandoned. -- Ambassador Delenn, "Babylon Five"

True to Warewulf's compartmentalized design, a number of subprojects come under the Warewulf project umbrella. Though most are modules of Warewulf itself, one that was recently added actually doesn't depend on the warewulf-common base at all! In fact, in the future there will be a Warewulf module which will rely on it.

The project is called the Warewulf Node Health Check, or simply "NHC" for short. NHC is an effort to build a framework, a tool, and a community around a common need: keeping track of whether or not compute nodes are ready to have jobs run on them.

Historically, most sites wound up writing their own one-off, home-grown solutions. While this approach is perfectly reasonable, understandable, and valid, it doesn't save anyone else the time and effort of having to recreate the same thing, nor does it leverage the prior work of others to the site's own benefit. So when faced with the same conundrum ourselves, we decided to take a different approach and write something that could be useful to a much broader range of HPC (and perhaps non-HPC as well) systems.

A lot of work has been done to fully document the script, the framework, and the pre-built tests it provides to make the learning curve as low as possible. Most recently, we've added tips on how to leverage built-in bash features to write your own custom checks optimally.

Give it a try, and let us know what you think! Feedback is welcome on the mailing list or in G+ comments. Enjoy!
Add a comment...
Wait while more posts are being loaded