Red Hat Cluster Suite
| Company type | Software suite |
|---|---|
| Industry | Computer software |
| Founder | Red Hat |
| Products | High Availability Add-On, Load Balancer Add-On |
| Parent | Red Hat |
The Red Hat Cluster Suite (RHCS) delivers software solutions designed for establishing high availability and load balancing clusters. These functionalities are provided via the High Availability Add-On and Load Balancer Add-On, both built upon open-source community initiatives, to which Red Hat actively contributes code upstream. Computational clustering is managed separately through Red Hat MRG.
High-Availability Add-On
The High Availability Add-On utilizes Linux-HA technology to maintain continuous service operation by overseeing cluster node health. A consensus among all nodes regarding configuration and shared service status is required to form a quorum, which authorizes service activation.
Communication of node statuses predominantly occurs over network interfaces (typically Ethernet), with alternative quorum approaches such as shared storage or multicast employed if the primary network fails.
This suite supervises and manages software services, filesystems, and network conditions, facilitating automatic failover to other nodes when failures arise.
To safeguard data consistency, fencing mechanisms—disabling a node's access to services or resources—are enforced through power disconnection (STONITH) or detachment from shared storage. Newer releases incorporate a distributed lock manager that provides granular locking capabilities without introducing single points of failure. Earlier iterations relied on the "grand unified lock manager" (GULM), which carried risks of failure and was last included in Red Hat Cluster Suite version 4.
Technical details
- Supports clusters with up to 128 nodes (limited to 16 nodes on Red Hat Enterprise Linux versions 3 through 6)
- Failover capabilities for NFS (Unix), SMB, GFS, and GFS2 file systems across various operating systems
- Service failover functionality
- Fully shared storage architecture ensuring robust data integrity
- Compatibility with SCSI and fibre channel technologies
- Includes OCF and LSM resource agents
Load Balancing Add-On
Red Hat’s load balancing component adapts the Piranha software to offer transparent load distribution and failover among servers. Applications do not require any special setup; incoming traffic is intercepted and directed by a Red Hat Enterprise Linux server according to predefined rules and metrics.
Support and Product Life-Cycle
Support provisions for Red Hat Cluster Suite correspond with the associated Red Hat Enterprise Linux release and its maintenance schedule. The product does not incorporate activation or remote disabling mechanisms, allowing continued operation beyond official support periods. Partial support is available for deployments on VMware virtualized environments.
History
The cluster suite has been included in the following releases:
- Red Hat Enterprise Linux 2.1 (released 2002-03-26)
- Red Hat Enterprise Linux versions 3.x, 4.x, and 5.x – featuring supported Global File System (version 1.x)
- Red Hat Enterprise Linux 5
- Versions starting from 5.3 introduced Global File System 2
- Red Hat Enterprise Linux 6.0 and subsequent releases, integrating Corosync (project), CMAN, and RGManager
- Beginning with version 6.5, options include Corosync combined with CMAN and RGManager or the Pacemaker Resource Manager
- Red Hat Enterprise Linux 7.0 onward (released 2014-10-06), incorporating Corosync alongside Pacemaker Resource Manager
See also
External links
- Overview of Red Hat Cluster Suite
- Managing and Configuring Cluster Fence Devices
- Guidelines for GUI-Free Configuration
- Red Hat Enterprise MRG Information
- Red Hat Enterprise Linux 5 Cluster Suite Overview
- High Availability Add-On Overview for RHEL 6
- RHEL 7 High Availability Add-On Overview
- Pacemaker Cluster Resource Manager Wiki
- CMAN FAQ on Fedora Hosted