Big websites around the globe like the New York Times, the UK Government and Reddit are coming back online after a glitch at a San Francisco tech company caused a massive internet outage early Tuesday morning.
Big outfits such as Amazon, Twitch, and Paypal all were affected by the disruption caused by South of Market-based Fastly.
The company acknowledged the problem just before 3 a.m. and about an hour later said the issue had been resolved.
That outage appeared to be caused internally.
People couldn't access key services and sites and instead received a 503 error which meant the server hosting the website was down or overloaded.
Why are so many websites dependent on so few Content Delivery Networks (CDNs) like Fastly?
“Fastly is one of the bigger providers of this,” Sam Sheckner, Wall Street Journal tech reporter, told KCBS Radio. “Their name implies what their goal is, which is to get the content to you more quickly than if you used a centralized server.”
He said websites use CDNs to load pages faster and protect against hackers.
“Security is actually one of the things that these CDNs often advertise. A DDOS, which is a Distributive Denial of Service Attack, is a lot easier to mount against a centralized server because you can overwhelm it more easily,” Sheckner said. “Websites often use these kinds of services because it makes them more resilient. But past web outages have been traced back to brakes and parts of the digital supply chain.”
It’s still not clear what caused this morning's glitch but Fastly said users may still see sluggish website loading times.