Is Amazon Still Down? AWS Outage Updates: What Websites Are Down Today and When AWS Will Be Back Up

Key Takeaways

• A large Amazon Web Services incident began in the US-East-1 region on October 20 and rippled worldwide.

• Amazon says the core issue has been mitigated and most services are back, though some users may still see intermittent errors while systems clear backlogs.

• Early indicators pointed to a DNS resolution problem that affected key services. Subsequent technical updates referenced issues in an internal EC2 network subsystem tied to traffic monitoring for load balancers.

• Consumer apps and platforms that saw disruption included Amazon’s own services, Snapchat, Venmo, Fortnite, Roblox, Reddit, Ring, Alexa, Duolingo, Canva, and more. Many education and enterprise tools, including Canvas LMS and Asana, reported issues as well.

• As of 1:30 a.m. on Tuesday, October 21 (Asia/Dhaka), the broad trend is recovery, with some pockets still stabilizing.

What Happened and Why It Matters

Amazon Web Services is a backbone for a huge share of the internet. When something goes wrong in a central AWS region, it can feel like the entire web is down. That is what many people experienced on October 20, when an incident in US-East-1 triggered elevated error rates and timeouts across popular websites and apps. Reports surged through the early morning hours in the United States, then rolled across Europe and Asia as the day progressed.

This was not a quick, isolated glitch. It affected login flows, payments, notifications, video streams, smart-home functions, and even background tasks that users never see. Because so many companies host their databases, queues, and microservices on AWS, a failure in one layer can cascade to many layers above it. The result is a familiar pattern during large cloud incidents: some sites go completely dark, others load slowly or time out, and a few work fine but can no longer process specific actions such as sign-ins, purchases, or content uploads.

Is Amazon Still Down Right Now?

For most users, no. Amazon states that the root issue has been mitigated and services are recovering. That said, recovery is rarely uniform. You might be able to browse Amazon while a friend in another region still sees occasional errors. Some applications recover instantly once upstream infrastructure is stable. Others must rebuild caches, rehydrate data stores, or process a backlog of failed jobs before everything feels normal again.

If your Amazon app still seems slow, this is typical in the hours after a major incident. Expect sporadic login delays, occasional throttling, or errors when starting fresh compute capacity while the platform clears queues and restores healthy baselines.

What Was the Cause?

Large incidents often begin as a specific technical fault that creates wider knock-on effects. Early signals around this event indicated a DNS resolution problem that interfered with calls to core services. Later technical explanations referenced an internal EC2 network subsystem tied to monitoring traffic across network load balancers. Those two descriptions are not mutually exclusive. Complex cloud platforms have many interdependent systems. A fault in one layer can present to customers as DNS failures, API timeouts, or resource launch errors even if the deeper cause sits elsewhere.

One important point for readers searching phrases like “Amazon hacked today” or “AWS shutdown”: there is no official indication that the event was caused by a cyberattack. The pattern, timeline, and remediation steps are consistent with a large-scale operational issue rather than a security incident.

What Websites and Apps Were Down Today?

Impact varied by region and by the way each service uses AWS. Based on user reports and company status notices throughout the day, the following services experienced disruptions or degraded performance at some point:

• Amazon-owned services: Alexa voice services, Ring, Prime Video, and portions of Amazon’s retail experience.

• Messaging and social: Snapchat and Reddit.

• Gaming: Fortnite, Roblox, and related stores or launchers for those ecosystems.

• Finance and payments: Venmo and several trading or crypto platforms reported issues tied to upstream AWS trouble.

• Productivity and SaaS: Asana, Airtable, Zapier, and other collaboration tools had periods of degraded performance.

• Education: Canvas LMS used by universities and colleges saw login and course-page errors reported by institutions and students.

• Creative and media: Canva, streaming services, and content platforms experienced intermittent errors or slowdowns.

This list is not exhaustive. Many smaller websites and regional apps were also affected, especially those that operate primarily out of US-East-1 or depend on services in that region.

When Will AWS Be Fully Back Up?

Amazon has said the underlying fault is fixed and that services are recovering. Full normalcy, however, takes time. Here is why:

1. Backlog processing: During a major outage, queued tasks fail or pile up. After a fix, those tasks must be retried. That load can make systems feel slow until the backlog clears.

2. Cache warm-up: Caches that were invalidated or lost must be rebuilt. Applications can operate while this happens, but performance is inconsistent.

3. Capacity rebalancing: Some teams pause deployments and scale operations during an incident. Restoring the pre-incident capacity plan can take several hours.

4. Regional variance: A fix applied in one region may propagate at different speeds. ISPs and DNS resolvers can also hold stale records for a while, which explains why one user is fine while another continues to see errors.

The practical takeaway: most users can expect normal behavior now, with occasional hiccups during the overnight and morning hours as systems continue to stabilize.

Was This a Nationwide Outage or a Global Outage?

It was global in effect, driven by a problem in a key US region that serves traffic worldwide. Modern apps rely on multiple AWS components across multiple regions. When a central region stumbles, dependencies elsewhere can break even if those other regions remain healthy. That is why users in different countries all reported trouble around the same time.

Why So Many Apps Went Down at Once

Search terms like “apps down right now” and “what websites are down right now” spiked for a reason. Cloud concentration is a feature of today’s internet. Many brands you recognize run on the same provider and, often, in the same default region. This creates efficiency and speed for developers, but it also introduces shared-fate risks. When a core AWS function misbehaves, hundreds of companies feel it. Engineers design mitigations such as multi-Availability Zone and multi-region architectures, but not every workload can fail over cleanly, especially for stateful services like databases, queues, or real-time analytics.

What To Do If Your Favorite App Is Still Not Working

Try these steps that commonly help after a DNS or control-plane incident:

• Log out and back in. Expired tokens often cause lingering errors after an outage.

• Power-cycle your network. Restart your router or toggle mobile data to refresh DNS and routing.

• Clear caches. In a browser, clear cache and cookies. In an app, force-quit and relaunch.

• Update the app. Some vendors push quick fixes during recovery.

• Check status pages. Most services post incident notes and restoration times.

• Wait a short while. If the problem is backlog-related, it can resolve on its own as systems catch up.

If you manage a business or campus service, consider pausing non-critical batch jobs, postponing large deployments, and monitoring queue depths while upstream recovery continues.

Lessons for Teams Running on AWS

Even if your product held up well, this event is a reminder to revisit your reliability posture:

• Region strategy: Confirm that critical paths are multi-AZ and assess whether true multi-region operation is justified for your workload.

• Dependency mapping: Create and maintain a clear map of external dependencies, including DNS, identity providers, queues, and storage.

• Failover drills: Regularly test DNS failover, degraded mode, and read-only modes.

• Back-pressure and retries: Ensure clients handle throttling gracefully, with exponential backoff and circuit breakers.

• Communication templates: Prepare customer messages in advance. Clear, fast communication reduces frustration during events like this.

FAQs

Is Amazon still down?
For most users, no. Amazon reports that the core issue has been fixed and services are recovering. Some users may still see intermittent errors while systems normalize.

Is AWS still down today?
The incident has been mitigated. Residual issues can persist for select services and regions while caches rebuild and backlogs clear.

What caused the AWS outage?
Indicators included DNS resolution problems and later references to an internal EC2 network subsystem that monitors traffic across load balancers. There is no official sign of a cyberattack.

What websites were down today?
Among the most visible were Amazon services, Snapchat, Venmo, Fortnite, Roblox, Reddit, Ring, Alexa, Duolingo, Canva, and Canvas LMS, with others reporting degraded performance.

Is Venmo still down?
Venmo reported disruption during the peak of the incident. Most users should now see normal behavior, with rare delays during ongoing recovery.

Is Canva or Canvas still down?
Both were affected. Canva and Canvas LMS functionality improved as AWS stabilized. A small number of users may still run into login or loading issues that resolve with time.

When will AWS be back up fully?
The platform is largely back, with full stabilization expected as backlogs drain and capacity rebalances. In most regions this should settle within hours, not days.

A significant AWS incident in US-East-1 caused a chain reaction that made it feel like the internet was broken. Amazon says the root issue is fixed and the network is healing. That matches the lived experience many users report as apps come back online. If something you rely on still feels flaky, try basic troubleshooting and give the platform time to finish recovery. For engineering teams, use this moment to review architecture choices, failover plans, and communication playbooks.