Test, plan, and prepare before disaster strikes

Datacenters, which are the lifeblood of any business, are responsible for managing and protecting more data than ever, and the effort to provide continuous access to this data and keeping applications running is one of the top challenges of IT managers.

The threat of disasters, whether man-made or natural, adds to that challenge, making disaster recovery planning (DRP) a critical need for every aspect of the IT infrastructure.

However, one large oversight that most companies have in common is the lack of testing of disaster recovery plans.

Why? Because testing disaster recovery plans is not easy. IT managers responsible for protecting data cite the lack of time and hardware resources as reasons for not running DRP tests. They also fear that testing will have a negative impact on the production environment itself - in effect, creating an IT disaster in the effort to avoid one. As a result, testing is often left out of the process or kept to a minimum.

To make matters worse, disaster recovery plans are not static. As weekly upgrades are made to the datacenter with modifications to the infrastructure, upgrades to the software, alterations to the storage to name a few, the thought of testing a disaster recovery strategy each and every time changes are made is cumbersome and unrealistic to most administrators, but that is exactly must be done in order to be fully prepared for any event.

But testing can have extreme advantages. Companies who include DRP testing get a better picture of the plan's effectiveness and have the ability to make changes to the plan before disaster strikes.

Knowing these challenges exist and the extreme advantages to mitigate the testing woes, VERITAS Cluster Server 4.0 and VERITAS Volume Replicator 4.0 software are fortified with new features that allow you to test, plan, and validate DR scenarios in production without disruption.

The VERITAS Cluster Server Simulator, the VERITAS Disaster Recovery Fire Drill and the VERITAS Volume Replicator Advisor, are all complimentary features that are key components of an integrated solution for high availability and disaster recovery.

Two of the three features - the VERITAS Cluster Server Simulator and the VERITAS Volume Replicator Advisor - are easy-to-use software tools that are now available for free download on VERITAS Architect Network. By downloading these tools you can experience the ease of use in planning ahead before disaster strikes.

VERITAS Cluster Server Simulator
VERITAS Cluster Server Simulator (see Figure 1) is a new stand-alone tool included with the release of VERITAS Cluster Server 4.0 and 4.1. VERITAS Cluster Server Simulator, which can be installed and managed on any laptop or desktop, allows cluster administrators to simulate application failover scenarios for planned and predictable recovery, based on application priority and server capacity. The simulator works exactly like the award-winning VERITAS Cluster Server, allowing you to test out what configurations would be right for your scenario.


Figure 1. VERITAS Cluster Server Simulator

By using Cluster Server Simulator, administrators can ensure critical applications are running on the most optimal servers in the cluster configuration in the event of a local or site fault.

Additionally, as multiple versions and complex application dependencies become more prevalent in a datacenter configuration, the Cluster Server Simulator can help ensure all pre-defined application dependencies are maintained in the event of a fault.

Cluster administrators can now freely test multiple application failover scenarios on any laptop or desktop and incorporate the saved changes into the production cluster environment to take effect immediately, this way administrators can verify new changes before implementation. The advantages of Cluster Server Simulator assist administrators in three ways:

1. Planning application failover scenarios - The Simulator verifies that applications are migrated to the most appropriate server based on planned failover strategy.

2. Testing various failover strategies - 'What if' application failover strategies can be tested on any desktop/laptop computer.

3. Maximizing server utilization - The Simulator can help an administrator better understand where certain critical applications are running and ensure that these critical applications are hosted on the best hardware in the cluster environment.

Since the Cluster Server Simulator can run on any laptop or desktop, the tool also provides an effective demonstration of the intuitive user interface of VERITAS Cluster Server, complete with wizards to guide you through setup of your first cluster.

Since the Cluster server Simulator takes advantage of the same graphical user interface that VERITAS Cluster Server uses, the administrator can have a practical understanding of how Cluster Server handles application failover either in a local area or over a wide area.

Note: The Simulator must be installed on a Windows 2000 or higher system. For a list of frequently asked questions, see the Simulator download page.

VERITAS Volume Replicator Advisor
The VERITAS Volume Replicator Advisor (see Figure 2) is a software tool that helps an administrator plan for optimal bandwidth when replicating data between sites.

By analyzing your environment over a two week period of time, the Advisor collects information on how much data is being written within your environment and analyzing this data based on different parameters. The Advisor uses the results of the analyzed data and gives you the optimal bandwidth recommendations and helps you configure your Storage Replicator Log (SRL) within VERITAS Volume Replicator.

The VERITAS Volume Replicator Advisor, which is now available by free download, provides the following key features:


• Performs both data collection and analysis


• Helps planning the SRL size after analyzing the sample data


• Provides the analyzed output in a graphical and text format


• Allows 'what if' analysis, which helps you to plan accurately and understand worst-case scenarios.

Note: The Advisor is a complimentary feature that is included with VERITAS Volume Replicator and is also available as a free download on VERITAS Architect Network.


Figure 2. VERITAS Volume Replicator Advisor

VERITAS Disaster Recovery Fire Drill
The VERITAS Disaster Recovery Fire Drill is the procedure for testing for the configuration's fault readiness by duplicating a failover without stopping the applications at the primary data center.

A typical Disaster Recovery Fire Drill brings up a database or application (on a data snapshot whether using VERITAS Volume Replicator or 3rd party replication solutions) on the secondary node to make sure that the application is capable of coming online on the secondary in case of a fault at the primary site.

In the event that the disk groups and application/database could not come back online, the administrator can review a log file that can help explain the issues so the administrator can take immediate corrective action. Disaster Recovery Fire Drill can be initiated anytime, without the need for extra hardware.

By using VERITAS Volume Replicator Space Optimized Snapshots, the snapshot of the data at the secondary site that is to be mounted for the application to come online uses minimal disk space (approximately 10% of what is currently utilized).

Once Disaster Recovery Fire Drill has brought the necessary applications online, the service group is then immediately taken off-line and verification of a successful failover is recorded in the log files.

The VERITAS Disaster Recovery Fire Drill is available as a complimentary feature within VERITAS Cluster Server 4.0 and 4.1.