Bird mess accumulating in ducts and aircon vents has contributed to data centre fires, according to the Computing survey. More predictably, perhaps, rodents are another scourge of IT systems, with numerous citations of cables gnawed by rats, rabbits and even squirrels.
Such incidents might have their funny side, but not when they cause electrical faults, fires and serious outages. They should certainly be considered in any disaster recovery strategy; after all, rats, mice, pigeons and other animals are commonplace in cities, particularly in old buildings, and are usually found in large, destructive numbers.
“Repeated squirrel attacks on our fibre backbone” said one respondent, while another said:
“Troublesome rodents in our old listed London City building gnawed through several network cables.”
Insects can also be a nuisance.
“A colony of bees decided the server would be a good place to nest, crawled into the power supply and jammed up the cooling fans."
Given the times we live in, acts of terrorism, while still unlikely to affect the average business, cannot be wholly discounted.
Indeed two respondents did report downtime due to terrorist-related incidents. One involved a member of staff who was arrested on suspicion of terrorist activities, the other a bomb going off in a neighbouring building. Fortunately no one was seriously hurt but the confusion caused by the ensuing police operation led to a power supply being wrongly connected and prevented access to the data centre which subsequently overheated and failed.
Any incident that is suspected to be linked to terrorism will, of course, be taken extremely seriously by the police, whether they eventually turn out to be serious threat or not.
“A man arrived in reception threatening to blow himself up. When the police arrested him they found that he had a string of sausages strapped to his waist.”
Off site and out of mind
While the survey reveals that one IT professional’s disaster is another’s bad day at the office, the difference lies in the degree to which the unexpected has been planned for. Even if organisations meticulously and frequently backup their mission-critical data and apps, their efforts will be wasted if the backup system resides in same data centre that has been damaged.
Risk mitigation is the name of the game and virtualised backups sitting on data centre hardware can be particularly vulnerable to sunny day thinking. Even though these virtual machines can be moved almost instantaneously from server to server, the fact that all the machines are in the same room is an obvious source of risk.
Instead, anything labelled “mission-critical” should be backed up or replicated offsite, preferably to multiple locations. Facilities range from a secure fire-resistant storage site for backup tapes, to specialised disk-based DR appliances, to fully replicated mirror sites, to hosted DR services in the cloud (Figure 5), depending on need and budget.
As long as mission-critical data and apps are safely copied to another location, then even if the power goes down, then the four horsemen will have to wait for another day.
Sometimes, the power of the mainframe is the most cost effective answer. Computing's Peter Gothard puts Computing's readers' questions on the future of the mainframe to IBM's Z13 expert Steven Dickens.
This Dummies white paper will help you better understand business process management (BPM)