The disruption that affected several websites such as Times of India, State Bank of India, and Swiggy, is not uncommon as a similar incident took place last year when AWS went down.
An outage at Akamai, the Content Delivery Network (CDN), took down large swathes of the internet on Thursday evening. CDNs are companies that websites small and large rely on to have their traffic distributed worldwide, and are an essential piece of the web’s architecture. In the Thursday incident reported by Akamai, the company cited an “issue with the Edge DNS service”. The Verge reported that affected services included the Times of India, Airbnb, Delta, and British Airways. Hindustan Times reported that websites and apps of the State Bank of India, Paytm, and Zomato were affected too.
Around one hour after the issue emerged, Akamai said at 5:09pm UTC (10:39pm IST), “This incident has been mitigated.”
What happened: DNS is a service that lets internet users’ devices map domain names to the actual webpages that they want to access. Edge services are essentially servers and associated processes that are located closer to users. In this instance, it seems like Akamai’s edge services malfunctioned everywhere and made it impossible for users to access webpages that are distributed from there.
Why this matters: The internet runs on CDNs, but little is known about the industry, which only has a handful of large players, including Google Cloud, Amazon Web Services, Cloudflare, and Akamai, among others. Any downtime at these companies has global ripple effects.
The incident was not a result of a cyberattack, the company said.
We are continuing to monitor the situation and can confirm this was not a result of a cyberattack on the Akamai platform.
— Akamai Technologies (@Akamai) July 22, 2021
What went down
Based on news reports (The Verge, TechCrunch, The Hindustan Times, TechRadar), here’s who was affected by the outage. Note that due to the high number of smaller websites that rely on CDNs, this list is far from a complete one:
- Times of India
- PlayStation Network
- Delta Airlines
- British Airways
- Southwest Airlines
- New York Metro Transportation Authority
- Fidelity Investments
- State Bank of India
- Fortnite (Epic Games)
- Xfinity (by Comcast)
- Xbox Live
- Electronic Arts (EA)
CDN downtime not unusual
It is not unprecedented for CDNs to go down like this, and take a huge part of the internet with them. Just last month, another CDN, Fastly, went down for an hour. Reddit, Spotify, Twitch, StackOverflow, GitHub, gov.uk, Hulu, HBO Max, and Quora were among the sites that were affected.
Last November, Amazon Web Services went down briefly as well, impacting internet in North America. That outage, per TechCrunch, impacted Adobe’s Creative Cloud, Roku, and even smart device ecosystems like the app-controlled autonomous Roomba vacuum cleaner.
Amazon Web Services is a key player in the CDN space. So much so, that in the aftermath of Akamai’s Thursday downtime, before network administrators around the world knew what was going on, the top post on a subreddit for reporting such outages speculated that such a large downtime could only be due to an AWS downtime: the title of the post was, simply, “AWS Outage?”
- Large Swathes Of Internet Go Down As Fastly CDN Faces Issues Globally
- CoWIN Portal Crashes Minutes After Registrations Open For Vaccination Of 18+
- WazirX Server Crashes As Trading Volumes Surge 2,000%, Token Up 1,000%