The theme of this issue is SUSE® Linux Enterprise 12. Articles highlight a couple of unique features in SUSE Linux Enterprise 12 and how to register SUSE products in the SUSE Customer Center, which debuted at the same time. The issue also includes Part 2 of a discussion with SUSE Vice President of Engineering Ralf Flaxa on the processes and methods SUSE uses to develop new releases and an update on certifications.

Happy reading!

Marjorie Westerman
Editor-in-Chief


Merger News

On November 20, 2014, Micro Focus International completed its merger with the Attachmate Group, formerly the parent company of SUSE. Going forward, SUSE remains a principal brand of Micro Focus. Learn more


SUSE Spotlight: Planning and Building a New SUSE Linux Enterprise Release, Part 2—A Conversation with Ralf Flaxa, Vice President of Engineering

As vice president of Engineering for SUSE, Ralf Flaxa is responsible for leading the team of engineers that develops SUSE solutions. He has contributed to the Linux community since late 1991 and has more than 15 years of international management experience building Linux products. In 2002 he joined SUSE where he has held various engineering management roles. He holds a Master’s degree in Computer Science from Friedrich-Alexander-Universität Erlangen-Nürnberg, Germany.

Editor’s Note: This is part 2 of a two-part series on how SUSE plans and builds new releases. Part 1, published in the June 2014 SUSE Insider and available here, focused on the phases of SUSE release planning and development.

Question: What are the processes and methodologies that Engineering uses for building a new release and why are they used?

In general, we chose a development methodology depending on the specifics of a product. This can be the classical waterfall methodology, the newer “agile” methods or a blend of both. A classical waterfall process is linear. You go through months and months of planning, and then you develop the entire release in clear development phases. With the start of the Beta Phase, all features are in it and then you try to find and fix all of the bugs and stabilize it. With these long phases, time-to-market is slow.

In agile development, you develop new features for two weeks and have something that works; then you add new features during the next two weeks and have something that works, and so forth. Because you are planning and developing in smaller chunks, you are improving a project every two weeks and always have a consistent, working status. Time-to-market is fast, and you can even pre-release the software to a customer who doesn’t want future features. With the agile method, if the market or your strategy changes, it’s faster and easier to adapt a new release versus with the waterfall model.

So it’s not surprising that most popular open source projects use the agile model, and SUSE uses agile development for our newer products, like SUSE Cloud and SUSE Manager. These are built in an agile way to be able to release them at any time and to integrate customer and, especially, early adopter, feedback and requirements in an ongoing way via continuous integration and updates. These are relatively new markets. Everyone is learning so there is a need to adjust as you go along.

However, in a large, complex project like SUSE Linux Enterprise Server, we need to interact and coordinate with many partners who have their own related internal planning (like scheduling the test lab or writing code). We also cover many hardware architectures and languages, which adds to the complexity. Also, since this is a well-defined, longstanding product, we don’t expect any dramatic requests or changes of direction. As a result, we use the waterfall model with its defined stages.

SUSE is known for its engineering excellence. With our mix of methods, we try to balance achieving the highest reliability for enterprise-class computing in our new releases while, at the same time, ensuring technical innovation as well.


Registering SUSE Products with SUSE Customer Center Using the Subscription Management Tool

Author: Sabine Söllheim is senior solution and product marketing manager at SUSE. With more than 20 years of experience in IT, she focuses on Systems Management solutions, Integrated Systems, SAP and special projects like SUSE Customer Center.

As part of the SUSE Linux Enterprise Server 12 and SUSE Customer Center launch, SUSE has provided a maintenance update to the Subscription Management Tool (SMT) for SUSE Linux Enterprise 11 SP3. This maintenance update enables SMT to register SUSE products with SUSE Customer Center and allows you to smoothly migrate from Novell Customer Center to SUSE Customer Center.

What does this mean for you? If you are running SUSE Linux Enterprise Server 12 and SMT, you need to update your SMT server(s) in order to connect to SUSE Customer Center. The update for SMT is available via the update channels. After the update, all your systems will automatically be connected to SUSE Customer Center. The updated SMT server(s) will continue to work as proxy server(s) for all SUSE products. And don’t worry, in all cases SMT can be rolled back to Novell Customer Center.

What if I have SMT 11 SP2 and earlier versions? Please update your SMT server to SMT 11 SP3.

How can I see the registrations of an SMT server? From the SMT server, open a shell window and enter the following: $> smt list-registrations

How do I get SMT to retrieve information and sync with SUSE Customer Center? From an SMT server, open a shell window and enter the following:
$> smt scc-sync

How do I configure SMT to work with SUSE Customer Center instead of Novell Customer Center? SMT can be configured either via YaST or the migration can be done via script. To change your SMT configuration from Novell Customer Center to SUSE Customer Center, run: $> smt-ncc-scc-migration

Where can I find more information on SUSE Customer Center and SMT?

Provide us your feedback via the feedback button in SUSE Customer Center.


Advanced Systems Management with Machinery

Author: Cornelius Schumacher is an engineering manager at SUSE leading the team responsible for the development of the Machinery tool. He has a long history working on systems management and build tools from products such as SUSE Studio™, the Open Build Service or YaST. Cornelius also is a long-time contributor to various open source projects.

Introduction
In SUSE Linux Enterprise 12, SUSE now offers Machinery for system administrators as part of its Advanced Systems Management module. Machinery is a new and unique systems management toolkit for Linux that supports configuration discovery, system validation and service migration. However, it does not replicate the functionality of existing tools such as SUSE Manager or configuration management systems. Rather, its purpose is to fill gaps and make integration of the existing tools easier, as driven by the needs of administrators in the data center. Here’s a closer look at what Machinery does and, if you are a system administrator, how it can help you with your job.

System Analysis
One important function of Machinery is system analysis. This is critical for many use cases you face in the data center. You need it for documenting system setups, validating correct operation of systems management tools, meeting compliance requirements or consolidating infrastructures that have grown. It is also a part of more complex work flows, such as migrating systems from physical machines to cloud-based infrastructure and a part of your QA strategy in a golden image approach. In addition, you can use it to feed your configuration management database.

The first step of an analysis is inspecting a running system. Machinery performs inspection without requiring any instrumentation of the system; it just relies on existing ssh infrastructure. The result can be shown on the command line or as a graphical report.

But Machinery doesn't stop with inspecting a system and creating a report based on this.

The System Description and Its Uses
The core concept of Machinery is a universal system description, which can be used in many different scenarios. The inspection is used for writing these descriptions, and Machinery provides you with the tools to manage them. Having this description of a system opens up a number of interesting possibilities.

You can, for example, compare different system descriptions. This is useful for validating if a system conforms to some reference. A system description can also be used to compare different systems or what has changed over time in the same system. Machinery also stores the descriptions, so you can even perform these activities without accessing any running system.

The system description can serve as a base for a migration as well. You can combine it with image building tools such as KIWI to create images for SUSE Cloud or other virtual or cloud infrastructures. Machinery can also be used to migrate systems from one version of an operating system to a newer one, for example, to take advantage of the new features of SUSE Linux Enterprise 12. Inspection and analysis provide insight into what needs to be adapted or what can be automatically migrated. Additionally, Machinery can be used to track and validate results of these migrations.

With the concept of the system description and a modular set of commands, Machinery provides an extensible set of features to cover a huge number of use cases.

Getting Started
To get started with Machinery, enable the Advanced Systems Management Module in YaST by calling yast2 scc and following the instructions there. Then run zypper install machinery to install the tool. Its man page contains a detailed description of what you can do with it. You can view it with man machinery. The command machinery help gives detailed help about the individual commands Machinery provides. One good way to start is to inspect an existing machine with machinery inspect <hostname> and show the result as a report in your web browser by calling machinery show <hostname> --html.

Machinery is developed as an open source project. Its code is available on GitHub, so you can follow and even contribute to its development. Regular releases are issued for SUSE Enterprise Linux 12 and made available through updates in the Advanced Systems Management module, where you get full support for it.


Introducing Wicked Network Management

Authors:
Karol Mroz is a developer at SUSE and a member of the team responsible for Wicked. Prior to joining SUSE, his time was spent working with SCTP (Stream Control Transmission Protocol) and MPI (Message Passing Interface), developing a network OS at Cisco, and developing video and teleconferencing software. He works out of a home office in Vancouver, Canada.

Olaf Kirch has been actively working on Linux for more than 20 years, covering a variety of areas such as networking, security and kernel development. Today he is a director of engineering, focused on the development of SUSE Linux Enterprise products.

Many people have very pragmatic interests when it comes to cars: they want their car to be reliable, big enough to accommodate their needs and not guzzling more gas than it needs to. Others take a great interest in what's happening under the hood, how powerful the engine is, and so on.

In the context of managing the network configuration of your SUSE Linux Enterprise Server installation, if you belong to the first group of people, you can stop reading here, SUSE Linux Enterprise Server meets your needs: it’s reliable, adaptable and rock-solid. However, if your technical interest in networking goes deeper, read on.

Wicked: What’s under the Hood
In a nutshell, what we did in SUSE Linux Enterprise Server 12 was to replace all of the bringing up, monitoring and shutting down of network interfaces, but without changing the user experience. You could say we took the car and replaced the engine, tires, gearbox - just leaving the body untouched. Oh, and the steering wheel is still vintage 11!

To maintain that familiar experience, we didn't want users to have to immediately memorize a new set of commands. Thus, you'll find the usual ifup, ifdown, ifstatus and ifprobe commands right where you would expect: in /sbin. Further, we made sure that your ifcfg files, and sysconfig in general, remained the default way in which you specify your network configurations. However, this is where the similarity ends, and Wicked truly begins.

Wicked has been designed and built from near scratch. What possessed us to do this? It's no secret that the network landscape has evolved dramatically. State-of-the-art data and cloud environments are highly dynamic, and, with software-defined networking and an arsenal of physical and virtual devices, massive configuration changes can be brought about with little effort from Joe and Jane Admin. These highly complex networks need a management framework that can react flexibly to large-scale change. Furthermore, since evolution drives on, the framework needs to be extensible in order to handle new device types, concepts and configurations. These were some of the key forces behind Wicked. But what fully loaded model would be complete without some fun features like device hot-plugging, architecture independence and a small footprint? And if all you need is for eth0 to get configured over DHCP, we have you covered, too.

Wicked utilizes a client/server model that allowed us to define standard facilities for things like address configuration that integrates well with the overall framework. We wanted to look at network configuration as a service we could provide and so set out to employ a layered approach, with different dbus services created to handle device-specific, as well as more common or shared functionality. Stacking these services gave us a powerful way to handle configuration at different levels of the networking stack. We chose C as our primary tool for writing this set of functionality, and included within, the kind of functionality provided by tools such as ifconfig, ip, brctl, vconfig and ethtool.

We're an open bunch, and so we even broadcast event notifications, allowing interested applications to discover when, for example, a network interface has been added to the system, an IP address has been added to an interface or routing has changed.

For certain system administration tasks, we created a framework for extending Wicked with shell scripts. This allowed us to use some existing tools for updating things like DNS settings and also gives an administrator the potential for extending Wicked to suit very specific needs. For even further flexibility, Wicked offers a new configuration file format that is rich and well structured. Currently, this format is used internally when parsing sysconfig-style configurations, but is ready for early adopters to use and will become the default in a future service pack.

All that's left is for you to take Wicked for a spin. Wicked is packaged with SUSE Linux Enterprise Server 12 and SUSE Linux Enterprise Desktop 12. It's also available for openSUSE starting with version 13.1. We invite you to use it and would value any feedback you have. You may contact us at: wicked-devel@opensuse.org.

Additional information about Wicked, including a basic how-to, source code and package locations and mailing list information can be found on our Wiki.


Certifications Update

Marjorie Westerman is a Marketing writer at SUSE. She edits the SUSE Insider and SUSE News.

SUSE certifies a wide range of products to make sure our customers have the innovative, best-of-breed solutions that meet their business and IT needs and leverage their existing IT investments. Here are some certification updates.

YES Certified Hardware
Server and workstation certifications for SUSE Linux Enterprise 12 are well underway, with the first such certification bulletins published on October 27, 2014. Of the 569 servers and workstations certified between August 1, 2014, and November 17, 2014, for  SUSE Linux Enterprise 11 SP3 and SUSE Linux Enterprise 12—393  of these were tested and certified for the new release. 
Certifications on servers and workstations on one or both operating systems include Cisco, Dawning Information Industry, Dell, Fujitsu, Hewlett-Packard Hitachi, Huawei Technologies, IBM, Inspur, Lenovo, NEC, Oracle, SGI, Wincor Nixdor, Positivo Informatica and Unisys. Certifications on both versions of the SUSE OS are ongoing.

Highlights
SUSE alliance partners Cisco, Dell, Fujitsu, Hewlett-Packard, IBM and Lenovo now have a substantial number of servers and some workstations certified to run on SUSE Linux Enterprise Server 12. Here are some of the major server families with members running on the new SUSE release.

  • Dell PowerEdge. Dell PowerEdge servers include advanced rack, tower and converged infrastructure platforms and are designed for the widest range of web, enterprise and hyperscale applications. These include unified communication and collaboration (UC&C) solutions for Microsoft Exchange, Lync and Sharepoint; business processing and decision support; High Performance Computing (HPC); and virtualized and cloud computing. For more information on the SUSE and Dell solutions, click here.
  • Fujitsu PRIMERGY and FUJITSU PRIMEQUEST. Fujitsu PRIMERGY systems provide powerful, flexible data center solutions for companies of all sizes, across all industries and for any type of workload. These solutions include expandable PRIMERGY tower servers for remote and branch offices, versatile rack-mount servers, compact and scalable blade systems, as well as density-optimized scale-out servers. With SUSE Linux Enterprise Server 12, SUSE also supports all features of the mission-critical PRIMEQUEST server, including Dynamic Reconfiguration. Fujitsu PRIMEQUEST servers embody the best characteristics of mainframe and UNIX server reliability, supercomputer high-performance and the cost and flexibility benefits of open systems. To learn more about SUSE and Fujitsu joint solutions, click here.
  • Hewlett-Packard ProLiant. The latest servers in the HP ProLiant portfolio span four architectures—blade, rack, tower and scale-out. They are optimized for convergence, cloud and software-defined environments, and feature new technology innovations that increase compute capacity; enable a software defined enterprise; provide faster setup, monitoring and firmware maintenance; and deliver improved performance. Learn more

For a list of specific certified hardware devices and the SUSE OS they run on, visit the Yes Certified hardware bulletin catalog here.

In This Issue

Join SUSE Conversations and
get free gifts.

Experienced user of SUSE products? Turn your knowledge
into Amazon gift cards!

Visit the new SUSE Shop

Buy on-demand training videos

Get up-to-the-minute SUSE news
by following us on:

View past editions of the SUSE News and SUSE Insider

Contact us at susenews@suse.com with questions or feedback.