Disaster Recovery for Homesteaders, Part 2

Adager Sponsor Message

November 2004

Disaster Recovery for Homesteaders

Know the scope of your agreements and select the right site to stay safe

By Paul Edwards

Second of two parts

[Ed. Note: Substantial portions of the first part of this article were adapted from a publication titled “The Disaster Recovery Guide” copyrighted by John Painter of Computer Solutions, Inc., which operates Minuteman Recovery disaster recovery services.]

The Systems Manager Notebook is vital to the proper management of any HP 3000 site and is part of the Disaster Recovery Plan. Every site should have one, because it contains critical hardcopy information to back up the information contained on the system. It is used to manually recreate your environment as a last resort. All parts of the notebook have to be kept current at all times.

Recovery Systems

There are several types of system agreements that can be utilized to get your business data processing operation functioning again. These are hot site, mobile system, cold site, and shared usage. The cost for each varies, as does the response time involved in being up and running.

Remember, these agreements are similar to insurance. You assess the risks and costs involved and make a proper business decision that makes sense for your company.

Hot site systems are maintained for you at a third party location. These systems mostly duplicate the hardware you currently have installed at your present company office. Proper communications and sufficient user access is part of the service. Because these systems are available on a very short notice, the time to be functional is quite rapid. The cost is the highest for this option, but provides the best possible recovery. This is the choice that is usually recommended, if you can afford it.

The mobile site system is a completely operational computer system in a self-contained trailer provided by a vendor that can be re-located to your company site or any other site you choose. Since the system can be readily accessed, testing can be easily done when required. The cost is usually less than a hot site. How close the trailer is staged to your recovery site will determine the amount of time required to be operational. A concern would be condition of the access roads to your recovery location after a disaster.

A cold site is a pre-determined location. This location must be properly prepared for a quick installation of a system and the accompanying equipment with power, communications, and user access already available. This location could be another company building, but you have to hope that the location chosen is not in the disaster area. This option will cost less on an ongoing basis, but will take much longer to bring on-line. The computer equipment may be a system you have purchased and stored for this purpose. Or, you may have a contract with a third party vendor to have a system available on short notice to ship to your location. Proper testing of your recovery plan is very difficult or impossible in this case.

Shared access is an agreement between two companies to allow processing on the production system of one company in the case of a disaster at the other company location. This is usually the lowest cost option, but there are many problems associated with this choice. Usually, companies have growth that doesn’t keep up with the capacity of their system. So, the available storage space and processing power is probably limited. Third party software required to process your applications may not be on their system. It would be disruptive to the other company to do periodic recovery plan testing. This approach is not recommended.

Communications and Backup

In a disaster situation, the communications will be your responsibility and have to be planned for well in advance. During a widespread disaster, most communications vendors will be so booked up with essential repairs that getting installation of equipment and new lines will be impossible. So plan accordingly. Communications can be designed to allow users to have access to the HP 3000 systems remotely with Virtual Private Networking, either from remote locations or even from their homes.

Proper backup procedures are essential to preserving the data integrity of the HP 3000 systems and providing a means to recover the system without loss of critical information. Full backups should be done as often as weekly and partial backups produced daily. The creation of MPE Customized System Load Tapes (CSLT) should be done at least twice a month usually with the full backup. Offsite storage of backup media at a facility that can provide the media on-call is a must.

A RAID storage system will provide mirroring of all of your system and data file drives. A failed disk drive can be hot-swapped on one of these RAID systems without any downtime. This is an inexpensive and highly recommended addition to your installation.

A technology called shadowing or continuous data replication can be used in addition to media backup and will keep the system environment of the main site in synchronization with a remote hot site. This will facilitate the rapid switchover to the alternate system with a minimum of downtime.

As you continue to homestead, you must periodically communicate with your HP 3000 third-party software vendors to ensure their licensed software continues to operate correctly; their support is still at the expected level; alternate means of changing any copy protection codes are available; and that the software company remains responsive to your communications around the clock.

Summary

The US Congress approved the Accounting Reform Bill or Sarbanes-Oxley Act to tighten regulation of independent auditors and make company officers more accountable for their conduct. Aside from requiring corporate officers to take greater responsibility for the accuracy of financial reports, SOX mandates that organizations understand the risks that may impact the financial reporting process.

A proper assessment of this risk environment would likely include lesser-known operational and IT risks resulting from inadequate disaster recovery or business continuity plans. Your company needs to have a complete plan in place now to comply with these directives and protect the business.

The most important part of a successful disaster recovery plan for homesteaders is the periodic testing and evaluation of the plan. This provides the opportunity to fine-tune the steps and procedures of the plan and discover any gaps that exist. Periodically, use the data stored offsite, continue to train all levels of personnel, test the communications links, reevaluate the results, and then modify the plan as necessary.

It is true that major disasters are rare. But minor ones can happen at any time. It is also true that no amount of planning and preparation guarantees a successful recovery. However, a properly prepared and tested disaster recovery plan does increase the odds of a company surviving a disaster.

Paul Edwards is founder of Paul Edwards & Associates, an HP 3000 consulting firm that offers a Web-based product to produce a complete disaster recovery plan and provides a variety of consulting services. A 28-year veteran of the 3000, chairman of the MPE Forum, former Interex board member, and an OpenMPE board member, Edwards can be reached at pedwards@gte.net or 972.242.6660.