When I woke up this morning, I fully expected the internet to, you know, work. Unfortunately, I and scores of others got no response in return. As it turns out, there was a major outage that, according to one source, involved half of the internet. The impact upon both my wife and I was immediate: Alexa was down, both in our living room and our bedroom. Her lights lit up, but there was no response. We had to revert to 1996.
Just a few minutes ago, we tried again and found her to be responsive in the living room, but not the bedroom. I rebooted her in the latter and though she lit up, there was no response. Alexa is one of the many major components that went down on us all…giggity. Reddit, Venmo, and Snapchat were among the many sites to flatline.
Hell, some people who use the Ring doorbell system found themselves to be affected as well.
What caused all of this mayhem? Amazon Web Services’ network in North Virginia, known as the US-EAST-1 network, failed. That region of Amazon’s network is the most critical.
Some media outlets chose to report that AWS runs close to 50% of the internet. But that number is, for many reasons, unrealistic to say the least. That region crashed, taking many of the most popular sites with it. It all went down, no pun intended, at about 2:20 am New York time. Amazon services went away too, which explains why Alexa was not working. I mean, we actually had to turn on the lights manually. Manually! In 2025!
Other sites that went down included government agencies within the United Kingdom were affected. Square and Duolingo, as well as Pokemon Go were down too. Amazon itself was impacted as well, with Kindle and Amazon Prime suffered outages. For a brief while, those trying to teleconference through Zoom found themselves out of luck.
Those who wanted to Hulu and chill were left with no other option than to turn on their actual television sets, assuming that people still have those in 2025. Users of other streaming services were bereft of entertainment.
United Airlines and Delta passengers were grounded.
Apple had a bite taken out of it, at least for a little bit. McDonald’s and the New York Times went down as well. People who wanted a ride home using Lyft? Stranded, unless they defected to Uber.
There are even some people who were left with no other choice than to walk away from their laptops or desktops, pick up an actual physical book and then read that book. That’s unthinkable to many! Thankfully, I was raised in an era where people had to actually hold paper books to read, but this new generation likely doesn’t even know what a book looks like. Think about the impact this had on them!
But one of the biggest networks that was brought to its knees was the XBox network, meaning that tons of gamers likely went full retard. Other gamers of the world must have been contemplating suicide as Fortnite and PlayStation went offline too. Maybe even World of Warcraft. It is unclear as to whether or not the most successful franchise uses AWS but there was no mention of downtime on many of the game’s X profiles.
Full disclosure here: I am a gamer, but I only use WoW, and I can assure you that if for some crazy reason I was up all night playing the game and I were impacted, I would have shrugged and gone to, you know, bed. But I would have loved to see what went on in gamers’ homes when they were impacted.
Those wanting late-night deliveries through Doordash were left to fend for themselves as the company’s platform died because of AWS’ outage. OpenAI and reCAPTCHA were down as well. The latter is a big problem because of the fact that the numerous websites that rely on the service to authenticate users were literally let hanging.
So what the hell happened? As you might imagine, an investigation is underway, but experts are already saying that one of two things happened: something internal went wrong, or it was a major cyberattack. The latter seems unlikely as it would take a crazy amount of resources to pull that off.
Several sources, including NBC News, are reporting that the former seems the likely culprit. But that’s one hell of an internal fault! Someone’s pants are going to burn before we’re through it. Multiple news organizations are reporting that Amazon’s EC2 network is to blame.
However, another component of Amazon’s network, DynamoDB, suffered a critical error and that’s apparently part of what caused the drama. What everyone seems to be agreeing on is that Amazon’s internal components are involved. In other words, this was not a DDOS attack nor anything else like that.
It is also possible that, since it happened in the middle of the night, some poor son of a bitch technician did something to the zone file. Such a thing is not unheard of: many years ago, someone working on the zone file over at Google managed to take down their entire network, taking Gmail and other services with it.
It could come down to something really simple but critical: DNS resolution. If Amazon’s DNS configuration were to be disrupted, then their load balancers would fail, causing their entire network to almost instantly crash. If that’s the case, however, the issue should have been resolved in minutes, not hours.
And, as expected, many people flocked to X to bitch and moan. Many people bemoaned the fact that Alexa became inoperable, so at least I’m not the only one.
I am sure that the sites that I mentioned were not the only ones to go down. Numerous banks and governmental entities likely found themselves down for the count. It’s not just major companies. Small businesses and affluent individuals use the services as well, so a lot of people were impacted by today’s outage.
Law firms found themselves unable to access their cases. Tons of small business couldn’t do the simplest of transactions. The list goes on and on. Lots of businesses and people have been impacted today, but if they didn’t have some sort of redundancy plan in place, then they kind of deserve what they got.
I am willing to bet that many web hosting companies were taken down as well, meaning that many smaller websites and blogs may have disappeared for a while. People are still, after many hours, talking about it on X. Some people are blowing it out of proportion, saying that every single website went down. Not so!
There are many bloggers out there, myself included, whose websites were not affected, at least not that I know of. If my site went down, then it did so in the middle of the night, so missed visitors wouldn’t be a major impact for me. My major traffic spike typically happens in the early afternoon.
Those who use Google’s infrastructure, mainly Blogger, were high and dry. I say this even as some people are reporting, without evidence, that Gmail, YouTube, Google Maps and other Google functions went down. I highly doubt that a behemoth such as Google would do business with Amazon.
As for the claim that half the internet went down, that’s hyperbole. The actual number of impacted websites and other services is closer to 30%, which means that 70% of the internet remained functional. So, if you put it into perspective, what happened today was not that bad of a thing, especially given the fact that, after four hours, things started coming back online. A four-hour outage really isn’t that big of a deal, especially given that it happened overnight.
It could have been worse is what I’m trying to say. I am certain that millions of people found themselves affected, but there is absolutely no way that any one thing could being the entire internet down other than something happening to the root servers, the servers that basically tell internet traffic where to go.
As of the time that this article was published, the internet was still recovering, even hours after the event. Here’s why. Today’s incident shows what can happen if a customer, whether corporate or private, puts all of its eggs in one basket. My hosting provider does not use AWS and as such, this site was not affected.
If my website had been directly impacted, there wouldn’t have been much impact as only 7% of my traffic hits my site during the overnight hours, at least when it comes to traffic from within the continental United States. I couldn’t care less about international traffic, so if I had been affected during the overnight hours, I really wouldn’t care.
It’s a little off topic, but visitors who hail from foreign countries visit from the United Kingdom, The Netherlands and Australia. Today’s outage might have prevented them from visiting, but the traffic volume is so low that I couldn’t possibly care less if visitors from those countries were unable to do so.
Sites such as mine don’t really warrant any sort of response when something like this happens, but I can only imagine the conversations that are going on at various major companies and app producers. I can almost hear the sounds of phone calls being made to other providers.
What happened today shows what happens when mission critical sites and apps depend upon one provider. Just about every datacenter out there has multiple networks, so if, say, Level3 goes down, then the affected datacenter can automatically switch to another provider without anything being impacted.
Clearly, choosing just one provider can yield negative results, especially when you rely solely on cloud-based hosting. Today absolutely was a wakeup call for the major websites and apps and many of them will either leave AWS altogether or will bring in a backup network. I do realize, however, that it might not be that easy to do so.
The fact that this happened overnight in the United States means that in our country, not many people were actually impacted as normal people go to bed at night and as such, much of our country slept through it all. However, AWS is a worldwide service. As we saw today, the United Kingdom’s governmental websites were taken down during the business day and I am certain that other European entities were affected as well.
In the coming hours or days, I am certain that we will find out precisely what caused today’s incident and we will hopefully hear what Amazon plans to do about it so that what happened today does not become the new normal. All it took for much, but not all, of the internet to go crazy was one company’s failure. One. That should be enough to alarm anyone with a brain.
I feel bad for some of those who were impacted. Think about it: today, people had to actually talk to one another. People had to actually go outside. Others had to use the phone to conduct business. Seriously, it had to have been absolute chaos for those people. I mean, talking to each other instead of enjoying the internet at its fullest power? That’s unconstitutional!
What about Alexa? As I stated, and as of the time that this article was published, she’s now working in our living room, but she’s still sick in our bedroom. This is a major issue, because for the time being, we have to use the living room Alexa. It’s a small inconvenience, yes, but it’s enough of an inconvenience to make my blood boil.
We rely on Alexa for a lot in this house. I use her for weather and time checks and for sexually harassing Google Home. In order to do these things, I temporarily have to get up off the couch and walk down the hall to the bedroom as if it were 1996.
That is what enrages me.