BlackBerry Connection > IT Edition > BlackBerry Enterprise Server v5.0 High Availability

BlackBerry Enterprise Server v5.0 High Availability

BlackBerry Enterprise Server v5.0 High Availability

Ensure maximum uptime for your deployment

Minimizing downtime is a critical task for any administrator. The new High Availability mode in BlackBerry® Enterprise Server v5.0 makes this process easier, allowing you to better control when automatic and manual failovers occur. In this edition of the BlackBerry Connection newsletter, we’ll explain the key facts about BlackBerry Enterprise Server v5.0 High Availability mode and describe some of its most important features.

Configuring BlackBerry Enterprise Server v5.0 in High Availability mode ensures maximum uptime for your deployment. By installing a standby server, you will have a system in place that’s ready to take over in case of either scheduled or unscheduled downtime. Though this sort of redundant architecture might seem complicated, it isn’t. No additional licenses are needed to set up a standby server, and setup can be completed in as few as three steps, making High Availability an easy option even for medium-sized organizations.

Getting Set Up

The core philosophy behind High Availability in BlackBerry Enterprise Server v5.0 is that there should be no single point of failure in the BlackBerry architecture. As long as your system is configured in High Availability mode, any single component of the system can fail without the entire system failing. (See figure 1.)

HA Diagram Figure 1

The idea of redundancy isn’t new—before BlackBerry Enterprise Server v5.0, there were several different techniques for ensuring maximum uptime for your BlackBerry architecture— but all of them involved manual intervention in the case of a failure, which takes valuable time and staff resources, increasing the cost of ensuring maximum uptime. With BlackBerry Enterprise Server v5.0, a robust system for automatically managing failover is available out of the box. (See figure 2.)

Simple BlackBerry Enterprise Server architecture Figure 2

High Availability mode can be as simple or as complex as your business requires. For medium-size operations with a basic BlackBerry Enterprise Server v5.0 setup, High Availability can be as easy as three quick steps, providing enterprise-grade reliability right out of the box. (See sidebar for details.)

Complex BlackBerry Enterprise Server architecture Figure 3

However, High Availability also scales up gracefully, ensuring that there is no single point of failover in even the largest and most complicated BlackBerry Enterprise Server v5.0 architectures. (see figure 3.)

As always, in-place upgrades are designed to be easy for all server configurations, and migrating your current fleet of smartphones to BlackBerry Enterprise Server v5.0 with High Availability is made even easier with the BlackBerry® Enterprise Transporter tool, which assists with phased rollouts. The BlackBerry Enterprise Transporter tool, which is a part of the BlackBerry Resource Kit, available here, allows system administrators to move BlackBerry® smartphones from one domain to another. Because the source and destination domains don’t have to be running the same version of the BlackBerry Enterprise Server software, you can use this tool to transition smartphones from an old BlackBerry Enterprise Server v4.x environment to a new BlackBerry Enterprise Server v5.0 architecture, without having to erase data or reactivate phones.

The tool can be used in Live mode, which allows migrations to take place without disrupting the operation of either domain, or in Bulk mode, which requires the source domain to be taken offline, but allows for rapid migration of large numbers of smartphones.


Automatic and Manual Failover

In order to ensure that there is no single point of failure, a BlackBerry Enterprise Server v5.0 architecture running in High Availability mode has a primary and a standby BlackBerry Enterprise Server, managed with a flexible system of automatic and manual failover procedures. The highly customizable nature of these failover conditions allows you to create a system that is perfect for your own architecture—providing resiliency in the face of hardware failure while avoiding unnecessary failovers or failover loops.

Automatic failovers allow you to reduce downtime by eliminating the need for human intervention to deal with hardware failure. Also, automatic failure detection means that you’ll never have to wait for a user complaint to act on a problem, ensuring short response times to problems.

Because failover thresholds are based on system health checks, the standby server can sometimes take over before the primary server actually fails, providing uninterrupted service. Customizable promotion thresholds allow you to specify a set of conditions that the standby server must meet before it can be promoted, so the system never replaces an unhealthy server with an even less healthy one.

Manual failover allows you to force a failover, which can be used to dramatically cut back on downtime during server upgrades. It allows you to first perform maintenance or an upgrade on the standby server and then manually initiate a failover, letting you upgrade the primary server without taking the entire system offline.

Health Checks

At the core of the BlackBerry Enterprise Server ability to provide smart, automatic failover is its capability to accurately gauge the health of each BlackBerry Enterprise Server. Health scores are pieces of data that are communicated between the primary and standby dispatcher. Each component calculates a health score, and these scores are then aggregated by a dispatcher to maintain a per-machine health level.

Each component running on a BlackBerry Enterprise Server keeps track of its own health (as determined by its operating efficiency, its connection to other components, and the health scores of the components it connects to), and reports that health to a single BlackBerry Dispatcher. The Dispatcher records the data from all health checks in a database and provides this data to BlackBerry Enterprise servers that request it, allowing the standby server to monitor the health of the primary server and to promote itself if its automatic failover thresholds are met.

Your organization’s BlackBerry Enterprise Server administrator sets two thresholds or you can use the out-of-the-box thresholds. The first threshold is the failover threshold, which sets the criteria for initiating a failover. The promotion threshold, which is for the standby server, sets the health criteria that must be met for the standby to become promoted. By adjusting these two thresholds, you can prioritize services.

To take a real-world example, you can choose what is more important, email or web browsing. You might decide that if web browsing breaks on the primary server and everything’s healthy on the standby, then that’s a good enough reason to failover. But if web browsing is not working on the primary server and email is broken on the standby, email is a more critical service, so you wouldn’t necessarily want to failover in that scenario. You’d be in, effectively, a more degraded state. So you could move the web browsing, or the MDS connection service below the promotion threshold, but leave it above the failover threshold. This system of health checks and thresholds lets you get a little bit more control over when you failover and why.

Learn More

This has been just a cursory look at the High Availability features available in BlackBerry Enterprise Server v5.0, but we hope it’s been enough to demonstrate that—with comprehensive health checks, smart automatic failover, and the powerful BlackBerry Enterprise Transporter Tool—these features will make ensuring high availability relatively simple.

For more information about BlackBerry Enterprise Server v5.0, visit the official website—and click here to download a free 60-day trial. For more information on high availability and management solutions from BlackBerry partners, visit the BlackBerry Solutions Catalog site.