Both major Linux players in the enterprise market -- SUSE and Red Hat -- have clustering tools, and each tool offers different capabilities. But there are some changes in the technology and direction tentatively slated for Red Hat Enterprise Linux 7 for 2013 that solution providers should be aware of when helping customers start out with a new Linux high-availability clustering tool.
High-availability (HA) clustering is a technique that can be used to make sure that vital business services are restarted immediately if something goes wrong. HA clustering is typically implemented when businesses run mission-critical services. The clustering service monitors that these services are still up and runningand makes sure they are restarted as soon as they stop functioning.
The history of HA clustering on SUSE and Red Hat
Red Hat has had the Red Hat Enterprise High Availability add-on for more than seven years, previously marketing it as the Red Hat Cluster Suite and optimizing the product along the way. The tool is now called the Red Hat Enterprise High Availability add-on in recent versions and is still the default clustering tool in Red Hat Enterprise Linux (RHEL) 6.3. Its development is directed by Red Hat Engineers.
More on Linux HA tools
Improve customer's vital services protection with operating system HA
Using STONITH to safeguard Linux HA server integrity
Tutorial: High availability options in a virtual environment
Meanwhile, SUSE has been one of the most important contributors to the Pacemaker project, an open source HA clustering project that evolved from the legacy Heartbeat project, which was launched in the late 1990s. SUSE has been working with developers from many different companies on the Pacemaker project, making Pacemaker the HA clustering solution used by many Linux distributions. The Ubuntu Server distribution replaced the Red Hat HA add-on with Pacemaker in the most recent Ubuntu Long Term Release version for the enterprise.
Pacemaker consists of a lower layer called OpenAIS/Corosync, and an upper layer referred to as Pacemaker. Both Red Hat's HA add-on and Pacemaker use the OpenAIS/Corosync layer in their solution. The main difference in current versions is found in the layer that manages services. This layer is called rgmanager on Red Hat and Pacemaker in Pacemaker.
The rgmanager tool provides needed services and manages the availability of these services, but it lacks certain features. Because of this lack of features, Red Hat has offered the Pacemaker packages as a nonsupported-technology preview version in the latest releases of Red Hat Enterprise Linux 6. As Red Hat moves toward adopting Pacemaker fully into RHEL 7, there will soon be no difference between HA clustering on SUSE and HA clustering on Red Hat. Because of this shift, support for the Red Hat HA add-on will likely end within the usual five-year window following the shift to Pacemaker, leaving customers in a lurch. Additionally, any future development on the Red Hat HA add-on will also cease, which means customers may face a lack of functionality in the short-term.
If customers need to create a new HA cluster, it's best to start with Pacemaker in either SUSE or RHEL. Pacemaker is fully supported in the SUSE High Availability Extension Pack and comes with comprehensive management utilities, including a Web-based management console and a graphical utility. Using Pacemaker in Red Hat currently requires management from the command line.
About the author:
Sander van Vugt is an author and independent technical trainer, specializing in Linux since 1994. Van Vugt is also a technical consultant for high-availability (HA) clustering and performance optimization, as well as an expert on SLED 10 administration.