--- title: "Amazon Web Services outage: The failure lasted over 12 hours, and issues with websites in multiple locations have yet to be resolved" type: "News" locale: "zh-CN" url: "https://longbridge.com/zh-CN/news/261941187.md" description: "Amazon Web Services (AWS) experienced a major service outage on Monday, causing numerous mainstream websites to crash, with the disruption lasting over 12 hours. Although some websites were restored within a few hours, fault monitoring platforms showed a surge in fault reports. Amazon stated that it is addressing the backlog of tasks and expects to complete the recovery within two hours. The outage affected the Amazon official website and several subsidiary businesses, with the main issue stemming from a DNS problem with its database service DynamoDB" datetime: "2025-10-20T21:20:41.000Z" locales: - [zh-CN](https://longbridge.com/zh-CN/news/261941187.md) - [en](https://longbridge.com/en/news/261941187.md) - [zh-HK](https://longbridge.com/zh-HK/news/261941187.md) --- > 支持的语言: [English](https://longbridge.com/en/news/261941187.md) | [繁體中文](https://longbridge.com/zh-HK/news/261941187.md) # Amazon Web Services outage: The failure lasted over 12 hours, and issues with websites in multiple locations have yet to be resolved As a leading company in the cloud infrastructure market, Amazon Web Services (AWS) reported a significant service outage on Monday, causing many mainstream websites to go down. Although some websites were restored within hours, the outage monitoring platform Downdetector showed a surge in reports of issues with Amazon, AWS, and Alexa (Amazon's smart voice assistant) around noon Eastern Time. In the latest update released by Amazon at 4:52 PM Eastern Time, the company stated that it is currently addressing a backlog of tasks and expects to complete the processing in about two hours. In a blog post on Monday afternoon, Amazon noted that AWS customers encountered an "increased error rate" when trying to launch new instances of its popular cloud service EC2. The company wrote, "We are working to restore services as quickly as possible." Around 1:30 PM Eastern Time, AWS reported that it had observed "initial signs" of EC2 service recovery in some regions and was implementing fixes for the remaining areas, stating that "once repairs are completed, instance launch errors and network connectivity issues are expected to be alleviated." Amazon also confirmed that the outage affected the Amazon website, some subsidiary operations, and AWS customer support services. The outage was first reported at 3:11 AM Eastern Time in AWS's primary region located in Northern Virginia — the US-East-1 region. Notifications on the AWS status page indicated that its database service DynamoDB was experiencing domain name system (DNS) issues, which is a core service supporting many other AWS applications. The role of the domain name system is to convert website domain names into IP addresses, allowing browsers and other applications to load content. In an update at 5:01 AM Eastern Time, AWS mentioned that this "operational issue" affected "multiple services" and stated that it was "accelerating recovery through multiple parallel paths." The outage impacted over 70 of AWS's own services. At 6:35 AM Eastern Time, AWS stated in an update that the DNS issues had been "fully resolved," and the operation of AWS services had "returned to normal." According to data from Synergy Research Group, AWS is the leading company in the cloud infrastructure technology sector, holding about one-third of the market share, ahead of Microsoft and Google. Millions of businesses and institutions rely on the cloud computing services provided by AWS, such as server and storage services. **Major companies affected** Downdetector shows that the reported malfunctioning platforms include: Disney+, Lyft, McDonald's app, The New York Times official website, Reddit, Ring smart doorbell, Robinhood, Snapchat, United Airlines, T-Mobile, and Venmo. In addition, Downdetector data indicates that the UK government website Gov.uk (UK government official website) and the HM Revenue and Customs official website also experienced outages. A government spokesperson stated, "We are aware of the issues with Amazon Web Services, and many online services relying on its infrastructure are affected. Through the established fault response mechanism, we are in communication with AWS, which is working hard to restore services as soon as possible." Lloyds Banking Group confirmed that some of its services were affected and requested customers to "please be patient during this time," while stating that it is working on service recovery. About 20 minutes later, the group added that services have been gradually restored. The outage also caused critical internal tools at Amazon to fail. Employees at Amazon warehouses, delivery personnel, and Flex service drivers reported on Reddit that internal systems at multiple sites were offline. Some warehouse employees were notified during their shifts to remain on standby in the break room and loading area; at the same time, they were unable to log into Amazon's Anytime Pay app — which allows employees to withdraw part of their wages instantly. The core platform "Seller Central" used by Amazon's third-party sellers to manage their businesses was also inaccessible due to the outage. A Reddit spokesperson stated, "We are currently working hard to restore Reddit services to 100% normal levels." Some passengers of United Airlines and Delta Airlines reported on social media that they were unable to check their booking information, check in, or check luggage online. A T-Mobile spokesperson stated that its users encountered issues accessing other websites or services due to the AWS outage, but the carrier itself "did not experience service interruptions or failures." The online teaching platform Canvas also stated that it was affected by "the ongoing AWS outage." Additionally, social media users reported that several cloud gaming services were down, including Roblox and Fortnite; cryptocurrency exchange Coinbase indicated that many users were unable to access its services due to the outage. Graphic design tool Canva stated, "The error rate has significantly increased, affecting all functionalities of Canva. This issue stems from a major outage at our cloud service provider (AWS)." The generative AI search tool Perplexity was also impacted. The company's CEO, Aravind Srinivasan, posted on the X platform, stating, "The root cause is the AWS outage, and we are working hard to resolve it." **The Hidden Dangers of Centralized Software** In recent years, large enterprises have been affected by technical failures more than once. In July 2024, a software upgrade error by cybersecurity company CrowdStrike exposed the vulnerabilities of global technological infrastructure—this error caused the Microsoft Windows system to crash, resulting in chaos worth millions of dollars, grounding thousands of flights, and disrupting the normal operations of hospitals and banks. AWS has also experienced multiple service interruptions in recent years. A failure in 2023 caused numerous websites to go offline for hours; a more severe failure in 2021 affected websites and services globally, including a temporary halt in some of Amazon's own delivery operations. For a long time, Amazon, Microsoft, and Google have been competing for enterprise customers. Earlier this month, after a service interruption in Microsoft’s office software suite, Google attempted to leverage this service failure to promote its own tools and launched a business continuity plan—running its Workspace service in parallel with Microsoft 365. In a blog post last week, Google wrote: “It’s only a matter of time before Microsoft 365 experiences a failure (the difference is when it happens and how long it lasts, not whether it will happen), but that doesn’t mean your team needs to revert to the days of pen and paper.” In June of this year, Google Cloud services experienced a prolonged outage that affected several major service providers, including OpenAI and Shopify. Google stated that this failure was caused by a series of recent flawed updates. Rob Jadin, Chief Digital Officer of cybersecurity company NymVPN, pointed out in a statement that Monday's AWS failure “does not appear to be triggered by a cyberattack, but is more likely due to a technical failure at one of Amazon's major data centers.” He added: “Such issues typically occur during system overloads or failures of critical network components, and since many websites and applications rely on AWS, the impact of the failure can spread rapidly.” When asked for comments, an Amazon spokesperson referred to AWS's service health dashboard. Mike Chapple, a professor of information technology at the Mendoza College of Business at the University of Notre Dame and a former computer scientist at the National Security Agency (NSA), stated: “Most consumers may not know what DynamoDB is, but it is one of the ‘core record systems of the modern internet.’” He further added: “In the coming hours to days, we will learn more details, but initial reports indicate that the problem did not lie with the database itself—the data appears to be secure. Instead, the issue was with the records used to inform other systems of the ‘data storage location.’” “This incident reminds us of how high the global dependence on major cloud service providers like Amazon, Microsoft, and Google is: when a mainstream cloud service provider ‘sneezes,’ the entire internet catches a ‘cold.’” ### 相关股票 - [Amazon (AMZN.US)](https://longbridge.com/zh-CN/quote/AMZN.US.md) - [Direxion Daily AMZN Bull 2X Shares (AMZU.US)](https://longbridge.com/zh-CN/quote/AMZU.US.md) ## 相关资讯与研究 - [Iridium Communications Stock (IRDM) Moonshots 12% on SpaceX IPO Filing and Amazon Takeover Rumors](https://longbridge.com/zh-CN/news/281548482.md) - [Should You Chase the Amazon-Driven Rally in Globalstar Stock Today?](https://longbridge.com/zh-CN/news/281573463.md) - [Why Is Globalstar Stock Soaring Thursday?](https://longbridge.com/zh-CN/news/281519853.md) - [Will Globalstar hit new highs following Amazon acquisition speculation?](https://longbridge.com/zh-CN/news/281533903.md) - [BREAKINGVIEWS-Satellite mania approaches $9 bln escape velocity](https://longbridge.com/zh-CN/news/281562888.md)