Although it’s encouraging to hear about perfect disaster recovery plans and outstanding recoveries, it’s also refreshing from time to time to hear about frank accounts of recoveries that weren’t so perfect. This isn’t for gloating. It’s for tempering optimism with reality and remembering that “the map is not the territory”. In other words, what you think might happen and what really happens are not necessarily the same things. Here’s what flummoxed ASCDI (Association of Service and Computer Dealers International), a worldwide association based in Florida.
The disaster that hit ASCDI and so many other organisations in America in 2005 was Hurricane Wilma, one of the biggest storms in the history of the United States. As the story posted by Joseph Marion on the ASCDI website (www.ascdi.com) states, ASCDI’s disaster recovery plan seemed solid. It allowed for 10 hours of UPS for the association’s servers, back-up generators after that and data backed up to a secure location at a remote alternate site as well. The association even went as far as providing access to the alternate site via a different web address. All apparently wise precautions for servers powering a website sending out critical information to computer resellers and over 250,000 emails a day.
What wreaked havoc on the effectiveness of the disaster recovery plan were several unforeseen factors. The back-up generators delivered “spikey” power that blew up equipment and left the main site out of action. Although new generators were delivered overnight, there were queues for hours at filling stations, restrictions on fuel purchases and the entire region had sold out of fuel cans. In addition, many ASCDI members were not aware of the address of the back-up site and continued, in vain, to try to access the main site. To find out what ASCDI learned and changes it made as a result, read the next post – “Business continuity plans and lessons learned”.