Disaster recovery is an important topic for network administrators — one that you should take seriously. What you need to know for the Network+ exam is the purpose and characteristics of disaster recovery processes. Expect to see at least one disaster recovery question on the exam.

Disaster recovery must be implemented when a natural or external event crashes the network and makes its resources, especially the data resources, unavailable to its users.

The purpose of disaster recovery is to restore the critical data that support the operational needs of the organization.

Here are some things disaster recovery isn't:

• Mistakenly deleting a file doesn't constitute a disaster, and restoring the file doesn't constitute disaster recovery.

• Routinely restoring data files from a daily backup to load the files on a new or modified server is not disaster recovery.

A disaster-recovery plan details how you restore the computing, application, and data resources of your network if any of these resources are lost, stolen, or destroyed. For example, a good disaster recovery plan includes preparing standard responses for

• Replacing a damaged network, server, or networking device.

• Replacing a damaged or failed network or server component, such as a power supply or disk storage device.

• Restoring a full data backup to a recovered or replaced server or network attached storage device.

Your organization should prepare and maintain a disaster-recovery plan that details what to do to restore networks and resources in the event of any conceivable event. Include minor events, such as a single corrupted file, in your plan, and prepare for the worst by setting up agreements with other companies or agencies so that your company can use its computers if your facilities are destroyed. And, of course, don't forget to include anything and everything in between.

The best recovery plan would be to keep disasters from happening in the first place. That's why arguably the most important part of a disaster-recovery plan is the procedure that details how the network and its resources are to be protected against loss. And second only to recovery is how files will be restored if they are lost. If your disaster-recovery plan includes agreements that allow you to use another company's computers in the event of an emergency, but you don't have a backup copy of the current data resources, there isn't much point to the recovery plan.

Although you need to know about most aspects of disaster prevention, one part of disaster preparedness that you don't need to worry about in detail is taking data backups. This is reserved for the CompTIA A+ and Server+ exams.

Preventing data disasters of the hardware kind

For the most part, hardware is very reliable these days, especially hard drives. But hard drives can and do fail, and the thought of losing everything, including the operating system, application programs, and system and user data from one fatal failure is the network technician's (and the user's) nightmare.

You can prevent data loss from hard drive failure. The most commonly used is RAID (Redundant Arrays of Independent Disks), which you will see many questions about on the exam. RAID stores data in multiple locations so that even in the event of a serious crash, all of the important data is still available. As cool as RAID is, however, it's generally not a good idea to put all your disaster preparedness eggs in one basket. You also need to protect against the failure of several other hardware components. Here are the ones you're likely to encounter in the situations that questions on disaster planning will pose:

• Power Loss: A loss of power at the wrong time can be devastating. Even if you use RAID, the data in RAM and even cache memory is lost when the power source is lost. An appropriately sized UPS (uninterruptible power supply) and, of course, a data backup are the best ways to protect against a power failure.

• Resource Conflicts: If two peripheral devices are set up with conflicting system resources, which include IRQs (interrupt requests), I/O (input/output) addresses, and DMA (direct memory access) channels, the conflicts can cause a network device and any data it's holding to be lost.

Living with Mother Nature

An essential element in any disaster plan must be your plans for dealing with natural disasters, such as a lightning strike, fire, flood, earthquake, tornado, mudslide, hurricane, avalanche, or locust swarm. Each of these natural disasters is capable of destroying your office, including its network servers and the data they contain. Bad hair days don't count as natural disasters.

The idea is that in a natural disaster managers need to deal with what are likely to be very important human issues while the network technician implements the data recovery plan to ensure that the business or organization can be operational with as much of its data as possible, as soon as possible.

Data disasters of the software kind

You should expect to see a question or two about software-related data disasters on the Network+ exam. Software-related data disasters are those caused by buggy software, viruses, and file system failures. Data backups are the best defense against software-related data loss. Don't sweat the details of data backups, but you should know their worth.

Rushing to disaster

Unfortunately, many software developers believe they are forced to push out their programs as fast as possible to stay ahead of the competition. The result is that software can be buggy and unpredictable. In fact, these days, buggy software is almost the norm. You may have heard the rule about waiting until the first service pack is released before installing new software versions. Definitely one of the top two reasons to have a solid data backup program is to handle the devastating effect of buggy software. (The other reason is hardware, if you couldn't guess.)

Most of the antivirus software programs you would use on your server include utilities that enable the software to keep itself (and all user workstations on the network) current automatically over the Internet. Virus attacks can be very malicious and result in severe data loss.

Remember that even the best antivirus software is no guarantee that servers and individual workstations are safe. Be prepared to answer at least one or two questions on the Network+ exam about antivirus software.

Protecting against corruption

It is important to keep the server hard disk in tiptop shape by performing regular maintenance and having the file system scanned on a regular basis. Keep an eye on the server's file system because the file structures that contain programs and data can become damaged. In some cases, this corruption can result in data loss.