Blue screens in offices, huge queues at airport check-in, delays at ports, blocked bank transactions, cancelled surgeries in hospitals. All this due to a bug in a small update to the Crodwstrike security software for Microsoft Windows.
It didn’t take an army of hackers. A single coding error was enough to cause scenes of chaos or paralysis of services from Britain and Germany to India and from Australia to the US. A full recovery will take time: days or even weeks, experts warn, in a development that reminds us just how dependent the world is today on digital services and a handful of tech companies.
Crowdstrike isn’t even among the giants. A company with a market cap of just over $80 billion is small potatoes compared to Big Tech, whose value is now measured in the trillions. But as one of the most popular cybersecurity companies on the market, its misstep has caused much of the world to stumble.
The main digital blackouts
The Crowdstrike – Microsoft disaster was the biggest IT system collapse in history. But this is not the first time we have seen our digital world change direction. There have been a number of such incidents in recent years. They peaked during the pandemic and lockdowns, when work, education and entertainment moved online. And as the parallel digital universe of our daily lives spreads, we will have to learn to live with it.
It is worth remembering the most shocking cases to date:
The Dyn Attack (October 2016)
From 2001 to 2017, Dyn had a strong presence in the online world, managing data traffic and providing DNS. When it was cyberattacked in 2016, all of its major customers – Twitter, Spotify, Netflix, Airbnb, Amazon, eBay, PlayStation Network, the Guardian, CNN and many more – went down with it.
The groups Anonymous and New World Hackers claimed responsibility for the attack, but it has never been confirmed whether they were actually the perpetrators. It was a DDoS attack, carried out through numerous DNS lookup requests from tens of millions of IP addresses.
The activities are believed to have been carried out through a botnet consisting of several internet-connected devices – such as printers, IP cameras and even baby monitors – infected with the Mirai malware. The Mirai botnet managed to take down thousands of websites simultaneously. It is estimated that this attack was twice as large as any other recorded to date.
When Amazon Web Services Went Down (March 2017)
A simple human error in March 2017 brought down the internet. According to Data Center Knowledge, an engineer was debugging when he typed a command incorrectly. That’s it.
A simple click of the wrong key crashed cloud services for several hours, causing a major headache for thousands of companies. It is estimated that the companies involved suffered losses of 150 million dollars.
75,000 British Airways passengers in the air (May 2017)
A computer system failure at IAG (BA’s parent company) led to the cancellation of hundreds of flights and left 75,000 passengers scrambling for alternatives. It was a major blow to the airline’s image. The official explanation was “a temporary power outage at one of our hubs”.
According to British media, the blackout was caused by a maintenance contractor who literally switched off a switch by mistake. The GMB trade union launched an attack on the company, stressing that the whole thing could have been avoided “if it had not laid off hundreds of dedicated IT workers and outsourced their work to contractors in India”.
What will happen now without Gmail and Google Drive? (December 2010)
At the height of the pandemic on December 14, 2020, when all communication and entertainment was digital, some of Google’s most popular services, including Gmail, Google Drive, and YouTube, went offline.
The outage, which affected hundreds of thousands of corporate email accounts, lasted just 45 minutes. However, so many people are so dependent on Google services that it was 45 minutes of…horror.
When Fastly Took Down the Media Giants (June 2021)
Reddit, Amazon, CNN, Bloomberg, PayPal, Spotify, Al Jazeera, and the New York Times were just a few of the internet and media giants affected when Fastly went offline for an hour. Hundreds of governments and other organizations also suffered digital outages lasting from minutes to an hour as a small programming error took down Fastly’s entire network.
It was a big deal, but it didn’t last long. And while the company was temporarily in the crosshairs, it was credited with not only reporting the issue immediately, but also ensuring it was fixed quickly. The fact that half the news on the internet went down that day showed just how popular Fastly’s service was, and how important the customers were to it.
The Akamai Failure (June 2021)
A week after the Fastly incident, the websites of dozens of financial giants and airlines in the US and Australia briefly went down on June 17, 2021 due to a failure on Akamai servers.
The company assured that it was not a cyber attack and fixed the error in about an hour, although the problems in Australia persisted for longer.
No Facebook, WhatsApp for 6 hours (October 2021)
2021 was a busy year for the digital world. A total of 7 significant blackouts were recorded (in the article we mention the 4 with the greatest impact).
On October 4, it was Meta’s turn to crash. People were left without Facebook, Instagram, WhatsApp, Messenger and Oculus Quest for about 6 hours. It was caused by what Mark Zuckerberg called a “nasty bug”, which occurred when “configuration changes in the backbone routers that coordinate traffic between our data centers caused communication disruption”.
It may not seem like a big deal to temporarily lose access to a social media and messaging platform, but Facebook also acts as an authentication mechanism for several other companies/services.
The problem was such that Meta employees had to manually reboot all systems. And because “anything that can go wrong, will,” for some time employees were unable to enter the building where the data centers are located to debug.
The entire incident cost Facebook $60 million in lost advertising revenue and $47.3 billion in market capitalization. Zuckerberg lost $6 billion of his personal wealth in the few hours that Facebook went down.
Another AWS ‘outage’ (December 2021)
The second Amazon Web Services (AWS) outage, after the one in 2017, occurred in December 2021. Along with AWS, Disney, Netflix and Spotify went down. Digital assistants Alexa and iRobot had connectivity issues.
Additionally, many colleges in the US canceled exams because they could not access the platforms that hosted them. On December 22, when its information systems collapsed, Amazon was still “recovering” from two major power outages earlier that month at one of its data centers.
Two hours without Spotify and Discord (March 2022)
“Something is wrong,” Spotify admitted in a tweet in March 2022, when it began receiving reports from thousands of users who were unable to connect to listen to their favorite music.
Around the same time, Discord also went down, joking in a similar post asking its users to “get out.”
The issue lasted for about 2 hours, as in addition to the separate minor issues the two services had, Google Cloud (Spotify and Discord’s service provider) announced that it had its own outage to address.
First Twitter, then Instagram (July 2022)
It was July 14, 2022, when two popular social media platforms went down within hours of each other.
Twitter was down for 40 minutes that morning. It was the third incident that year (two others preceded it in February) due to a bug, which caused anger among its users.
A few hours later, Instagram went down, with some losing access to the platform entirely and others unable to send DMs. The irony? Instagram users took to Twitter to share their ordeal, which lasted about two hours.