A major internet outage on Tuesday happened after a single customer changed configuration, which unleashed a previously undiscovered bug, revealed Fastly, a US-based company that runs a content delivery network (CDN) used by the affected websites.
CDNs enable customers to store data, such as videos and images, on a large pile of "mirror services."
“We experienced a global outage due to an undiscovered software bug that surfaced on June 8 when it was triggered by a valid customer configuration change,” Nick Rockwell, Fastly's senior vice president of engineering and infrastructure, argued in a Tuesday’s blog post.
Among the websites that were affected by the outage were Bloomberg News, the Guardian, the New York Times, Reddit, and Twitch, with users also reporting trouble getting access to YouTube, Twitter, and Spotify.
According to Rockwell, the problem was discovered “within one minute,” prompting technicians at Fastly to disable the ill-fated configuration. The vice president said that in less than an hour 95 percent of the network was back to normal.
“This outage was broad and severe, and we’re truly sorry for the impact to our customers and everyone who relies on them,” Rockwell wrote.
Apparently the bug was included in May’s software update and the company is now trying to figure out why it was missed during trials.
The Tuesday glitch prompted some newsmakers, namely the BBC, to move away from Fastly’s network at the moment of outage. It’s unclear whether other clients will follow suit in the long run, but the situation highlighted the over-reliance on one service provider. The UK’s government website was also affected by the bug.
