↓Skip to main content

ClusterLabs projects

The cluster stack

  • The Open Cluster Framework (OCF) is a set of standards for cluster components. Currently, only the resource agent standard is in use. A resource agent is an abstraction that provides service-level awareness to a cluster. It contains the logic for what to do when the cluster wishes to start, stop, or check the health of a particular service.

  • The Quite Boring Library, also called libqb, provides high-performance features for client/server applications, most importantly logging and interprocess communication.

  • Kronosnet, also called knet, is a network abstraction layer supporting redundancy, security, fault tolerance, and fast fail-over of network links.

  • Corosync is the membership and messaging layer. It tracks all hosts in a cluster, including liveness and quorum, and allows for reliable communication among hosts, with message deliverability guarantees.

  • Pacemaker is the cluster resource manager. It coordinates the configuration, start-up, monitoring, and recovery of interrelated services across all cluster hosts. Pacemaker supports a number of service types, including OCF resource agents and systemd unit files. Pacemaker supports advanced service configurations such as groups of dependent resources, services that should be active on multiple machines at the same time, resources that can switch between two different roles, and containerized services.

  • The fence-agents project provides a set of commonly used fencing agents. Fencing agents are the abstraction that allows Pacemaker to isolate badly behaving nodes, by either powering off the node or disabling its access to common resources. The fence-agents project provides fence agents for devices including intelligent power and network switches, IPMI, cloud providers, virtualization hosts, and shared storage access.

  • The resource-agents project provides a set of OCF resource agents for commonly used services.

High-level configuration tools

Pacemaker’s internal configuration format is XML, which is great for machines but terrible for humans.

Multiple higher-level configuration tools are available to allow the configuration to be viewed and updated in a more human-friendly format.

  • The crm shell is the original configuration shell for Pacemaker. Written and actively maintained by SUSE, it may be used as an interactive shell with tab completion, for single commands directly on the shell’s command line, or as a batch mode scripting tool.

  • The High Availability Web Konsole (Hawk) is a web-based GUI for managing and monitoring ClusterLabs clusters. The Hawk API server provides a minimalistic web server handling SSL certificates, proxying, and static file serving for Hawk.

  • The Linux Cluster Management Console (LCMC) is a GUI with an innovative approach for representing the status of and relationships between cluster services. It uses SSH to let you install, configure, and manage clusters from your desktop.

  • The Pacemaker/Corosync Configuration System (PCS) provides a command-line tool pcs and the PCS Web UI for managing the complete life cycle of multiple cluster components, including Pacemaker, Corosync, QDevice, SBD, and Booth.

Cluster add-ons

  • The Anvil! Intelligent Availability platform is a self-contained system based on a two-node high-availability cluster with power fencing. Striker is the current, second generation of the platform. Anvil is the upcoming, third generation of the platform. The anvil-external-deps project is collection of scripts and spec files used by the platform.

  • The Booth cluster ticket manager extends Pacemaker to support geographically distributed clustering. It does this by managing the granting and revoking of tickets which authorize one of the cluster sites, potentially located in geographically dispersed locations, to run specified resources.

  • The cluster-glue library initially provided an executor and fencer for early Pacemaker versions that did not include their own, but now mostly provides Pacemaker the ability to use fencing agents conforming to the Linux-HA interface.

  • The ha-cluster-exporter project provides a Prometheus exporter for monitoring ClusterLabs clusters.

  • The PostgreSQL Automatic Failover (PAF) project provides a featureful OCF resource agent for PostgreSQL.

  • SBD provides a self-fencing mechanism for cluster hosts, utilizing a watchdog device and optionally shared block storage such as a SAN, iSCSI, or FCoE.

Miscellaneous

  • The clusterlabs-www project provides the source for this website.

  • The go-pacemaker project provides Go language APIs for cluster configuration and status.

  • The High Laughability project is an occasionally amusing collection of anecdotes (mostly chat quotes) from ClusterLabs software development.

Historical projects

The following projects are no longer maintained and are provided solely as a historical reference.

  • The fence-virt project provided fencing agents for virtual hosts, now included in the fencing-agents project.

  • The Pacemaker 1.0 repository provides access to early Pacemaker code.

  • Pacemaker-mgmt was the first GUI written for Pacemaker.

  • The nagios-agents-metadata project provided support files for using Nagios monitoring plugins as Pacemaker resources (no longer supported as of the Pacemaker 3.0.0 release).

  • The SAP Cluster Connector provided integration tools for managing SAP software in a cluster.