That outage impacted millions of people and large portions of the internet around the world, with — according to various published reports — Facebook, Reddit, Robinhood, Venmo, Verizon, Lyft, United Airlines, Ring, Snapchat, The New York Times, Fortnite, Roblox and the McDonald’s app among the sites whose users encountered problems.
Users began reporting problems starting around 3 a.m. eastern standard time, according to Downdetector. By 5:30 a.m. on the east coast, Amazon said its systems were recovering and many apps and sites had come back online.
“The underlying DNS issue has been fully mitigated, and most AWS Service operations are succeeding normally now,” AWS wrote in its update. “Some requests may be throttled while we work toward full resolution.”
AWS continued providing updates through the day, reporting recurring issues with EC2 instance launches, API errors and network connectivity.
Just before 7 p.m. on the east coast, AWS said on its Service Health page that the issues had been resolved, though some services have some catching up to do.
Advertisement: Scroll to Continue
“Over time we reduced throttling of operations and worked in parallel to resolve network connectivity issues until the services fully recovered,” AWS said in an update posted at 6:53 p.m. on the east coast. “By [6:01 p.m. on the east coast], all AWS services returned to normal operations. Some services such as AWS Config, Redshift and Connect continue to have a backlog of messages that they will finish processing over the next few hours.”
The company added in the update that it will share “detailed” post-event summary.
Earlier in the morning, the company said it was “investigating increased error rates and latencies for multiple AWS services” in a region of data centers in the eastern U.S., concentrated around Northern Virginia. AWS added its engineers were “engaged and are actively working on both mitigating the issue, and fully understanding the root cause.”
The company said the issues were the result of issues with the Amazon DynamoDB system, which provides database storage and computing power to websites.
As The Wall Street Journal noted in a report on the outage, AWS, the largest cloud-computing service provider in America, supports millions of websites and platforms, which means any problems with its network can affect the wider internet, including popular mobile apps.
In other AWS news, PYMNTS spoke recently with Nilesh Dusane, the company’s global head of Institutional Payments, about the way payments have become a competitive differentiator for financial institutions.
“Our customers are moving from the ‘lift and shift’ model that they used to use for cloud … to truly build cloud-native payment applications on AWS,” Dusane told PYMNTS during a discussion for the September “What’s Next in Payments” series, “From Trend to Table Stakes: Mapping the Next Payment Priorities.”
This marks a potential inflection point. There was a time when banks and payment companies adopted the “lift and shift” approach to which Dusane referred to the cloud by moving existing applications away from legacy data centers and onto the cloud to harness efficiency and scalability. The last few years, however, have seen a pivot to cloud-native design.
“The conversation, as a result, is no longer just about time-to-market but about time-to-value,” PYMNTS wrote.
