High Availability Messaging Solution Using Axigen, Heartbeat And Drbd

By Author: AXIGEN Team
Total Articles: 18
Comment this article

This white paper discusses a high-availability solution for the AXIGEN Mail Server, using the Heartbeat package with DRBD. AXIGEN is a proprietary messaging solution while both Heartbeat and DRBD are open source software released under the GNU Public License (GPL). These software packages can be used together to build scalable and highly available integrated cluster messaging applications on the Linux operating system.

Products Overview
AXIGEN (http://www.axigen.com/mail-server/isp/) - Leveraging on proprietary technologies SmartProcessingâ„¢, UltraStorageâ„¢ and GrowSecureâ„¢, the AXIGEN Mail Server delivers redundancy at the mail protocol level of a messaging solution, as well as for the actual message container storage.

AXIGEN is a carrier-class messaging solution with advanced email filtering and routing capabilities, also featuring groupware and advanced collaboration functionalities, and backed up at the same time by around the clock, highly effective technical support.

Heartbeat (http://www.linux-ha.org/Heartbeat) - A fundamental part of the High-Availability Linux project, Heartbeat ...
... provides core cluster management services, including membership, communication, resource monitoring and management services, IP address takeover etc. Heartbeat version 1.2.3 supports multiple IP addresses and a simple two node primary/secondary model.

When used with AXIGEN in a cluster environment, multiple Heartbeat pairs, each of them consisting of two nodes, can be configured in order to support larger clusters with the additional help of a layer 4-7 load balancer, a LDAP centralized user database and, of course, AXIGEN proxy nodes. With the newly-released version 2.0.0 of Heartbeat, the two node size limit of the cluster is raised.

DRBD (http://www.drbd.org/) - It is a block device or disk replication technology that can be viewed as network RAID-1. It can be used effectively to mirror a whole block device via a network onto another block device. Thus, DRBD involves two block devices, one labeled as primary (local) and the other labeled as secondary (remote/backup/standby).

Every write operation to the primary local device is written to disk and also sent to the other host across the network to be written to the secondary device. The remote host (secondary) writes the data to its configured disk. If the primary node fails, then the secondary node can take over in a typical failover scenario.

Solution Architecture & Implementation
In the solution examined, DRBD can be viewed as a resource that is controlled by Heartbeat. This ‘node management' layer is transparent from AXIGEN's point of view. DRBD uses the underlying physical disk as a virtual device. This virtual device is essentially a DRBD resource that is used by AXIGEN to access the disk. When the DRBD resource is made ‘active' on a particular node, it means that the disk configured as a DRBD resource is now accessible and ready for I/O operations.