If you tried to launch Snapchat, Canva, or Fortnite this morning without success, you’re not alone. Since the early hours of Monday, October 20, a massive outage has shaken the global digital infrastructure, rendering many services inaccessible or unstable. The cause: a failure within the cloud environment ofAmazon Web Services (AWS)a little-known but omnipresent technological pillar of the Web.
Of Snapchat has Canvasvia Fortnite, Duolingo, Roblox, Signal, Perplexityand even operators like Orange Or Vodafonethe consequences of the outage extend well beyond the United States. And the first returns show that France is not spared.
AWS, the “invisible cloud” behind your everyday apps
The problem originates from the AWS region US-EAST-1located in Virginiathe true global heart of data centers. This is where Amazon centralizes an essential part of its cloud services, used by technology giants as well as start-ups, banks, public services and entertainment platforms.
The incident affects several critical services of AWS, including:
- Amazon DynamoDB (cloud database)
- Amazon EC2 (virtual servers)
In a message published at 9:51 a.m., Amazon technical teams stated: “We are actively working to mitigate the issue and understand the root cause.” This same message evokes “high error rates” and “abnormal latencies”.
According to Downdetectorwhich records outage reports in real time, thousands of users in the United States and Europe have reported malfunctions, mainly on mobile and on the web versions of many popular platforms.
Games, networks, professional services: a global domino effect
What is surprising about this outage is not only its severity, but its extent. Indeed, services of very different natures are simultaneously affected, a sign of their dependence on the same cloud provider.
Here are some examples of applications reported as partially or completely inaccessible:
- Snapchat : blocked messages, inability to connect
- Canvas : design tool remains offline for many users
- Fortnite And Roblox : game servers unreachable
- Signal And Perplexity : encrypted messaging and search engine disrupted
- Duolingo : problems with synchronization and opening lessons
- Coinbase : inaccessible for some, but the company reassures: “All funds are safe.”
In parallel, British banksas Lloyds (28 million customers), would have been briefly impacted, according to SkyNews. Even the telecom operators like Orange or Vodafone would have experienced slowdowns related to background cloud services.
What next, what risks for users?
At this stage, no cyberattack is mentioned. It would rather be a technical error or internal malfunction on a particularly busy server cluster. This type of outage, although rare, is feared because it affects millions of users in cascade, sometimes without them knowing that Amazon is involved in the operation of their favorite application.
AWS confirmed on its status page that it is still experiencing “increased error rates and latencies” on several services. And to add: “We are actively engaged and working to both mitigate the problem and understand the root cause.”
For users, the main consequence remains the temporary inability to access their usual serviceswithout any real danger for their data. On the other hand, for businesses, this outage is a reminder of the fragility of a sole dependence on a cloud provider.
According to the latest information, a gradual return to normal is underwaybut no precise timetable for resolution has yet been communicated.