Did the Internet Break Yesterday? For Millions, the Answer Was Yes.
For a few frantic hours on October 20, 2025, it felt like the internet had simply broken. Your Ring doorbell wouldn’t connect. Your kids couldn’t log into Roblox. You couldn’t order a Big Mac on the McDonald’s app. Even your Alexa smart speaker was giving you the silent treatment. For millions of people around the world, the digital services that underpin their daily lives had vanished into thin air.
This wasn’t a solar flare or a coordinated cyberattack. It was something far more mundane, and far more terrifying: a single point of failure in the heart of the global internet. The culprit? Amazon Web Services (AWS), the world’s largest cloud provider, and specifically, its sprawling US-EAST-1 region in Northern Virginia.
I’ve spent my career advising companies on technology risk, and I can tell you that what we saw yesterday was the predictable, and perhaps inevitable, consequence of a decade of cloud consolidation. We have built a global digital infrastructure that is breathtaking in its power and scope, but it is also perilously fragile. The AWS outage of October 2025 is a wake-up call, and it’s one that C-suite leaders cannot afford to ignore.
What Happened? A Post-Mortem of the Great Outage of October 2025
The outage began around 3:11 a.m. Eastern time, a time when most of the US was asleep, but the digital world was very much awake. The first signs of trouble were reports of increased error rates and latencies in the US-EAST-1 region. Within minutes, a cascade of failures had begun, taking down a who’s who of the internet.
The list of casualties is a testament to how deeply AWS is embedded in the fabric of our digital lives. Snapchat, Roblox, Fortnite, Robinhood, Coinbase, Signal, Ring, Alexa, Kindle, the McDonald’s app—all of them went dark. Even Amazon’s own e-commerce site was affected. In total, 64 internal AWS services were impacted, a staggering number that reveals the complex interdependencies within the AWS ecosystem.
The cause of this digital carnage? Not a sophisticated cyberattack, but a humble DNS issue. For those not versed in the technical jargon, the Domain Name System (DNS) is the phonebook of the internet. It’s the system that translates human-readable domain names (like www.google.com) into the IP addresses that computers use to communicate with each other. When DNS breaks, the internet breaks with it.
Recovery began approximately three hours after the outage started, but for many services, the effects lingered for hours. The slow and bumpy recovery is a stark reminder of the complexity of these systems. When you’re operating at the scale of AWS, you can’t just “turn it off and on again.”
The Ripple Effect: When One Region Goes Down, the World Feels It
The most alarming aspect of this outage is that it originated in a single geographic region: US-EAST-1. So why did a problem in Northern Virginia cause a global internet meltdown? The answer lies in the history of the cloud and the economics of data.
US-EAST-1 is the oldest and largest of AWS’s regions. It was the first region that AWS launched, and for many years, it was the default region for any company that wanted to build on the cloud. As a result, a huge number of services are still hosted there, and many of them have deep-seated dependencies on the region’s infrastructure.
I once advised a fast-growing startup that had built its entire platform on US-EAST-1. When I asked them about their disaster recovery plan, they told me they had a “multi-AZ” (Availability Zone) strategy. In other words, their application was spread across multiple data centers within the US-EAST-1 region. I had to explain to them that this was not a disaster recovery plan; it was a redundancy plan. If the entire region went down, as it did yesterday, their business would go down with it.
This is the systemic risk of cloud concentration. We have allowed a single company, and a single geographic region, to become a single point of failure for a significant portion of the global internet. This is not a criticism of AWS; they have built an incredible platform that has enabled a wave of innovation. But it is a criticism of the industry as a whole for failing to take the risks of cloud concentration seriously.
The C-Suite Case for True Multi-Cloud Resilience
For years, the conversation around multi-cloud has been focused on cost optimization and avoiding vendor lock-in. But the outage of October 2025 has made it clear that there is a far more compelling reason to adopt a multi-cloud strategy: resilience.
The financial impact of yesterday’s outage will be measured in the billions of dollars. For companies like Roblox and Fortnite, which make their money from in-game purchases, every minute of downtime is lost revenue. For financial services companies like Robinhood and Coinbase, an outage can erode customer trust and lead to a flight of capital.
The bottom line is that for any business that relies on the internet, a single-cloud strategy is no longer a viable option. The risk of a catastrophic outage is simply too high. C-suite leaders need to be asking their technology teams some hard questions:
- What is our true level of cloud dependency? Are we reliant on a single provider, or even a single region?
- What is our plan for a regional or even a global cloud outage? Do we have a true multi-cloud strategy that will allow us to fail over to another provider in the event of an outage?
- What is the financial impact of an hour of downtime for our business? And what are we willing to invest to mitigate that risk?
These are not easy questions, but they are essential ones. The cost of building a true multi-cloud strategy is not insignificant, but as we saw yesterday, the cost of not having one can be far, far greater. So where do you start? It begins with a comprehensive audit of your application portfolio to identify critical workloads and their dependencies. From there, you can begin to design a multi-cloud architecture that prioritizes resilience, starting with your most critical applications. This might involve using a combination of different cloud providers, or even a hybrid approach that combines public and private cloud resources.
The Hard Truth About Multi-Cloud
Of course, a multi-cloud strategy is not a silver bullet. It comes with its own set of challenges and complexities. Managing applications across multiple cloud environments can be a significant operational burden. It requires a new set of tools and skills, and it can lead to higher costs if not managed carefully.
There is also the risk of data gravity. As you start to store more data in a particular cloud, it can become increasingly difficult and expensive to move that data to another provider. This is a form of vendor lock-in, and it’s one that can be difficult to escape. The egress fees charged by cloud providers for moving data out of their network can be a significant barrier to a true multi-cloud strategy. This is not just a technical challenge; it’s a financial one. It requires careful planning and a deep understanding of the cost implications of data migration.
But these challenges are not insurmountable. With the right tools and the right expertise, it is possible to build a multi-cloud strategy that is both resilient and cost-effective. The key is to go into it with your eyes open, and to be prepared for the challenges that lie ahead.
We’ve Been Here Before: A History of AWS Outages
Perhaps the most frustrating aspect of this latest outage is that it is not a new problem. We have seen this movie before. AWS has experienced major outages in 2017, 2020, 2021, and 2023. Each time, the industry has wrung its hands and talked about the need for greater resilience. And each time, the conversation has faded away as the memory of the outage has receded.
This time needs to be different. The outage of October 2025 was not a minor blip; it was a systemic failure that exposed the fragility of our global digital infrastructure. It is a warning that we cannot afford to ignore.
The Future is Multi-Cloud
The era of the single-cloud strategy is over. The AWS outage of October 2025 has made it clear that the risk of cloud concentration is simply too great. For any business that wants to survive and thrive in the digital age, a true multi-cloud strategy is no longer a luxury; it is a necessity.
The transition to a multi-cloud world will not be easy. It will require new tools, new skills, and a new way of thinking about infrastructure. But it is a transition that we must make. The future of the internet depends on it. The question is no longer if you should diversify your cloud presence, but how and when. The answer, as yesterday’s events have shown, is now.