Amazon Web Services suffered a severe downtime Tuesday, the company said, cutting access to many popular websites.
The company provides cloud computing services to many governments, universities, and corporations, including The Associated Press.
Amazon said in a post an hour after the outage began that it had identified the cause and “was actively working towards recovery.” The problem primarily affected his services in the eastern United States, it said. She did not provide any further details about the cause.
Amazon later updated the dashboard and noted that the company was “beginning to see some signs of recovery. We currently do not have an ETA for a full recovery. “
The outage also impacted Amazon’s ability to provide status updates, it said.
Problems started this morning on the US east coast, said Doug Madory, director of internet analytics at Kentik Inc, a network intelligence company. “AWS is the largest cloud provider and us-east-1 is their largest data center, so any disruption there will have a major impact on many popular websites and other internet services,” he said.
Customers trying to book or rebook trips with Delta Air Lines have had problems connecting with the airline. “Delta is working quickly to restore functionality to our AWS-supported phone lines,” said spokesman Morgan Durrant. The airline apologized and urged customers to use their website or mobile app instead.
Dallas-based Southwest Airlines said it switched to west coast servers after some airport-based systems were affected by the outage. Customers reported DownDetector, a popular clearinghouse for user outage reporting, still outages, more than three hours after launch. Southwest spokesman Brian Parrish said there had been no major flight disruptions.
According to DownDetector, people who tried using Instacart, Venmo, Kindle, Roku, and Disney + have reported problems. The McDonald’s app was also down. However, American, United, Alaska and JetBlue airlines were not affected.
Toyota spokesman Scott Vazin said the company’s eastern US region for dealer services has declined. The company has apps that handle retailer warranty claims and other services that have affected over 20.
Madory said he didn’t think the failure was anything nefarious. He said a recent accumulation of outages among providers hosting large websites reflects the evolution of the networking industry.
“More and more of these outages are the result of the automation and centralization of administration,” he said. “This ultimately leads to failures that are difficult to completely avoid due to operational complexity, but are very effective when they occur.”
Kentik saw traffic to Netflix fall 26%, among the top web-based services hit by the outage, Madory said.
It was unclear how or if the failure affected the federal government. The US agency for cybersecurity and infrastructure security said in an email response to questions that it was working with Amazon “to understand the possible impact of this outage on federal agencies or other partners.”
Subscribe to Newsletter
Stay up to date with all the latest from Boston.com