Large sections of the internet went down on Tuesday morning.
Media publications including The Guardian, the Financial Times, The New York Times and ZDNet, as well as websites including Reddit, Twitch, Amazon, PayPal, and the UK government website gov.uk, went down due to an outage.
Visitors to the sites were met with an error message: "Error 503 Service Unavailable". The problem was widespread and was linked to an outage at cloud platform and content delivery network (CDN) Fastly.
Fastly's service status indicated that it was experiencing global CDN disruption, which caused the global issue, with services across North America, South America, Europe, Asia/Pacific, South Africa and India all suffering from a "degraded performance".
"We're currently investigating potential impact to performance with our CDN services," Fastly said in a service update posted at 09:58 UTC.
Services began returning about 45 minutes later when Fastly said, "The issue has been identified and a fix is being implemented" – at the time of publication, it isn't certain what the exact issue was, although the company has been contacted for comment.
Update: at 12:09, Fastly confirmed that a "service configuration" was the reason for the outage.
"We identified a service configuration that triggered disruptions across our POPs [points of presence] globally and have disabled that configuration. Our global network is coming back online," the company said via Twitter - and later confirmed in an email to ZDNet.
"Fastly has observed recovery of all services and has resolved this incident. Customers could continue to experience a period of increased origin load," Fastly said in an additional update at 12:41.