Man v the Internet: classic cases of humans crashing the web
24th August 2018
To err is human, to disconnect is disaster. Just the sort of thing the 18th-century poet Alexander Pope might have tweeted were he alive today and hooked up to social media.
Vodafone haven’t confirmed the cause of the outage that shutdown Gatwick Airport’s display boards early this week but we do know it was a damaged cable and we also know that there’s just shy of a 1-in-4 chance that it was down to human error.
According to the Ponemon Institute, a research centre dedicated to privacy, data protection and information security policy, the three most common causes of Internet outage are: power supply failure (25%); cybercrime (22%); and accidental human error (22%).
With those stats to hand, we’ve had a look through the archives to dig out some classic cases where man conquered machine, without even thinking, and collapsed connectivity across vast swathes of the population.
The fragile nature of our reliance on all things webby was perhaps best underlined when the nation of Armenia, population 2.9 million, lost all Internet connection back in 2011. The cause? An electromagnetic solar storm? Nope. A raging off-the-Beaufort-scale hurricane? Nope. A 75-year-old woman digging for scrap metal? Got it in three! To make matters worse for the marooned Armenians, the spade-wielding OAP was based in neighbouring Georgia.
Old MacDonald had a farm and on that farm he had a tractor, E-I-E-I-Oh dear it seems his plough has cut through vital fibre-optic cables. To compound the cut cable on South African farmland, that shutdown one of the major supply lines to Zimbabwe, the back-up line was damaged in a separate incident. The severed cables were owned by Liquid Telecom, which has an 81.5 percent market share of Zimbabwe’s international-equipped internet bandwidth, and saw much of the country’s tech industry limp to a halt.
Middle East & Asia 2008
“Anchors aweigh”, cried the captain across the deck of a ship floating off the coast Alexandria. “I said, ‘Anchors aweigh’”, repeated the captain. “We’re trying”, bawled back his crew, “it’s snagged on something.” The captain paused for thought, then remembered his paymasters back on land. “Pull harder, pull, pull, pull.” SNAP. “Anchors aweigh, cap’n.” And with that large tracts of the Middle East and Asia were without Internet access after a key undersea cable was damaged. Mahesh Jaishanker, an executive director for Du, the UAE’s largest telecoms company, said, “The submarine cable cuts affected at least 60 million users in India, 12 million in Pakistan, six million in Egypt and 4.7 million in Saudi Arabia.”
Amazon’s S3 cloud service experienced an outage of several hours that brought down some of the world’s most popular websites and apps, including Medium, Business Insider, Slack, Quora and Giphy. This time there wasn’t a tractor, anchor or spade in site. It all boiled down to a simple tpyo. The poor engineer, who chose to study a number-heavy degree for a reason, accidently took the wrong servers in the wrong quantity offline and in doing so caused a whole host of major issues. According to one estimate, it cost American companies over $300 million in lost revenue.