- High Availability Application Clustering for Linux

Production Ready

Download [1.2]
Download [1.4]
Developers Site [1.4]


Introduction is a generic toolset for helping deploy standard applications in a highly available manner. In summary it allows two separate machines to host the application. Each application is active on a single machine at a time, and if that machine fails, or loses all network connectivity the application is automatically started on the alternative machine.

Of course using two machines for providing a single access to a single application may sound expensive. However does not limit the number of applicatinos that can be run as separate entities on the cluster. Both nodes can be active running application simultaneously providing fail-over facilities for each other. aims to allow administrators to deploy standard applications very quickly and strives to be easy to setup and manage. It requires no special hardware - just standard PC's can be used. It does not require any 'shared storage' - each machine has a separate copy of the data replicated via the product automatically over a standard TCP/IP connection.

  • Ease of Deployment - available in many standard package formats with very few dependencies.
  • Ease of Setup - Just a single XML file is needed to define the cluster. Each application then has a separate configuration file as requried - also in XML.
  • Ease of Administration - a single command is used to build the cluster and one other to define an application.
  • Flexibility - Supports applications that use either file systems and raw logical volumes.
  • Virtualisation Support - a command allows virtual machines as applications to be live-migrated between nodes.
  • Dynamic Software - Once the cluster is running there is rarely a need to stop it. Applications can be added, changed and removed whilst the software continues to run.
  • Automated Recovery - If a node fails and the adminsitrator brings it back into the cluster it's copy of the data is automatically resynchronised - no actions are required by the administrator.
  • Intregrated Network Management - provides local IP fail-over to redundent networks connections if available.

Licensing and Further Information. is free software, available under the GNU GPL v2. If you have any further questions or queries please email me at simon.edwards <at>

Find Useful?
If you are able to consider making a donation to support on going development.
So how do I Start?
Have a look at the "resources" link above! On that page an installation guide, a how-to and step-by-step guides for using common applications can be found. If there is an application that is not documented which you are considering clustering please contact me simon.edwards <at> - and I will consider doing a step-by-step guide for it.

What do you use for?

I would like to here from anyone using - please let me know about your experiences - good or bad. The aim is to put together a page indicating how various companies make use of it. If you have any comments please send them to Simon.Edwards <at>

What applications have been successfully clustered?
The software is generic in nature and thus allows virtually any application that provides a service via an IP address, to be clustered. Applications that the software is currently use for include:
Apache Mysql PostgreSQL
OpenLDAP Samba Bacula
BIND Courier MTA NFS Server
Oracle 10g Xen Virtual Machines  

Last updated: 25/03/2011
Latest Production Ready DRBD 0.7 based release: 1.2.9
Latest Production Ready DRBD 8.2/8.3 based release: 1.4.10
Latest Development Release N/A