An outage at Cloudflare, a content delivery network used by many companies, took down several websites around the world on Tuesday.
The effect was felt by users of some of the major internet traffic hotspots, such as Twitter, Amazon Web Services (AWS), Discord, Zerodha, Shopify and Canva, according to Downdetector, an online platform that provides users with real-time information. overview of problems and outages on various websites and services.
Websites like Udemy, Splunk, Quora, Crunchyroll were also down, along with crypto exchanges like WazirX, Coinbase, FTX, Bitfinex, and OKX. Most of these websites were then accessed.
Cloudflare said it suffered an outage that affected traffic in 19 of its data centers, adding that “unfortunately, these 19 sites handle a significant portion of our global traffic.” “Even though these locations represent only 4% of our total network, the outage affected 50% of total requests,” he said in a blog post.
The outage was caused by a “change that was part of a long-term project to increase resilience in our busiest places”.
The outage began at 11:57 a.m. IST. At 12:28 p.m. IST, the first data center was brought back online and at 1:12 p.m. IST, all data centers were online and functioning properly, the company said.
Over the past 18 months, Cloudflare has worked to convert all of its busiest sites to a more flexible and resilient architecture. So far, it has converted 19 of its data centers to the architecture internally called Multi-Colo PoP (MCP). These are: Amsterdam, Netherlands; Atlanta, USA; Ashburn, USA; Chicago, USA; Frankfurt, Germany; London, UK; Los Angeles, USA; Madrid, Spain; Manchester, UK; Miami, USA; Milan, Italy; Mumbai, India; Newark, USA; Osaka, Japan; Sao Paulo, Brazil; San Jose, USA; Singapore; Sydney, Australia; and Tokyo, Japan.
An essential part of this new architecture is an additional layer of routing that creates a mesh of connections. This mesh allows Cloudflare to easily disable and enable portions of the internal network in a data center for maintenance or to troubleshoot an issue, the internet infrastructure company said.
On June 13 this year, Amazon.com’s e-commerce services were down for thousands of users, according to Downdetector.com.
Last year there was a major global internet outage on June 9, which Fastly, the company behind the problem, said was caused by a bug in its software that was triggered when a of its customers changed its settings. The outage had raised questions about the internet’s dependence on a few infrastructure companies. Fastly’s problem eliminated high-traffic sites, including news providers such as The Guardian and The New York Times, UK government sites, Reddit and Amazon.com.