October 21, 2025 • 9 mins
ABC News tech reporter joins the show for ‘Tech Tuesday.’ Today, Mike talks Amazon Web Services outage that affected dozens of companies and what they are doing to prevent it from happening again.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
It's Tech Tuesday with Mike Dubuski, ABC News reporter and
tech maven out of New York.
Speaker 2 (00:07):
Mike, good morning, Thank you for joining us.
Speaker 3 (00:09):
Yeah, good morning.
Speaker 2 (00:10):
How are you all right?
Speaker 1 (00:11):
The big news is the AWS meltdown, and it is
absolutely enormous.
Speaker 2 (00:18):
And so here's my first question. Because they are known.
Speaker 1 (00:23):
I think they're the world's biggest company that does cloud computing.
Speaker 2 (00:28):
They are.
Speaker 1 (00:29):
And you know, I'm very confused because I thought your
information in the clouds were in the clouds and that's
not how it works, is it. And if you could
explain how this actually works and then what happened and
what's going on right now?
Speaker 2 (00:45):
Yeah.
Speaker 3 (00:45):
Sure. So if you're a small business, let's say you
want to open a cafe in downtown Los Angeles or
what have you, it doesn't really make economic sense for
you to spin up like a computer server to run
your payroll or to run your payments processing or whatever
else it is you do online. It kind of makes
more economic sense for a lot of companies to outsource
(01:06):
that work to someone who knows what they're doing, you know,
a Microsoft or a Google, or in this case, an Amazon,
which accounts for the largest chunk of the cloud computing service.
This is a service that works, you know, ninety nine
percent of the time. The reason that Amazon accounts for
about thirty seven percent of this market and Google and
Microsoft account for another third of it goes back to
(01:28):
the idea that they run really robust systems that don't
go down ninety nine percent of the time. Well, yesterday,
in the case of Amazon, we saw that one percent
of the time crop up when an issue with specifically
their DNS cropped up. This is a domain name service. Basically,
to make a very technical thing kind of simple, it
is a piece of technology that translates human language into
(01:50):
something that a computer can understand. It's what you type
into a web browser being translated into an IP address.
So that is the sort of nature of the problem.
We don't know specifically what went wrong with it, but
that's what went down, and it created these sort of
knock on effects throughout the day. Because Amazon Web Services
is so big and it hosts so many different you know,
servers and technologies out there, you break one thing, it
(02:12):
actually ends up breaking like four or five things. So
there was like a kind of knock on effect, a
whack a mole situation going on where Amazon engineers fix
the main problem pretty quickly and then you know, realize
these later problems down the line. The good news is
that most things are resolved at this point. A quick
look at down detector right now finds that most of
the major players who are experiencing issues yesterday among them
(02:34):
then most Snapchat, Microsoft three sixty five, Google It excuse
me a zoom, I should say not Google, but like,
these are all systems that are now running relatively normally,
and you know, the people are somewhat back to normal.
You're a Starbucks order on the app should work this morning.
Speaker 1 (02:50):
Now, when you say thirty percent of the market that
Amazon has, is that just the American market?
Speaker 2 (02:56):
Are we talking internationally?
Speaker 3 (02:58):
It's the global market?
Speaker 2 (02:59):
And to it.
Speaker 3 (03:00):
Yeah, I was talking to a friend of mine who
is a tech journalist in the UK, and they actually
experienced this in a much more acute way because they're
ahead of us. This outage actually happened for you guys
kind of in the overnight hours. For us here on
the East coast, it was just after three am, and
most of America is asleep at that point, so the
sort of initial wave of problems were only noticed by
people who were up at that time. Of course, we
(03:22):
started to see more reports trickle in, you know, nine o'clock,
ten o'clock, as people go to work and figure out,
you know, hey, this thing I had access to I
don't have access to anymore. Well, in the UK, that
hit their government systems, right, the website that controls their
equivalent of the IRS went down in part because of this,
So you know, it really does underscore exactly how much
influence Amazon has in this particular market. We saw similar
(03:45):
stories come out of continental Europe and beyond in Australia
as well.
Speaker 2 (03:49):
I mean, it is astounding how big this is. Now.
Speaker 1 (03:54):
If I'm not mistaken, this AWS accounts for more money
and more prime than the retail end of Amazon itself.
Speaker 2 (04:02):
Do I have that right? It does. Yeah.
Speaker 3 (04:04):
We often think of Amazon as like this e commerce website,
that's where they make their money. Of course, that's the
public facing component of their business. But when you actually
look at their profits, the majority of them come from AWS,
come from this cloud computing service. Amazon is a cloud
computing company before it's an e commerce retailer. That's the
long and short of it. And you know that again
(04:25):
is something I don't think a lot of people understand it.
Speaker 2 (04:27):
It's really important episode.
Speaker 1 (04:28):
Now, this cloud servicing, it has to do with these
massive forms of computers that keep my information, your information's
process things. I'm almost thinking that the world of computing
is growing far much quicker than their ability than people
(04:53):
like the WS or companies to deal with it and
the amount of energy that it takes.
Speaker 2 (04:58):
You want to talk about that.
Speaker 3 (05:00):
Yeah, So I think that there are certainly protocols that
companies like Amazon can implement to make sure that they're
not shipping out faulty code that breaks things like we
saw yesterday, or maybe they raise the standards in the
coding for you know, how they deal with third parties
and what have you. We also, in recent years have
seen generative artificial intelligence play a bigger role in sort
(05:20):
of like entry level coding, in fact taking the place
of entry level coders at many tech companies. However, humans
are imperfect, right, They're going to ship out improper code
every once in a while, and so is generative artificial intelligence.
Despite the fact that we talk about it all the
time as this transformative technology, it's still pretty bad at
a lot of things and hallucinates and makes mistakes frequently.
(05:43):
And that means that in the same way that you
come across a typo in the newspaper every once in
a while, or your order at Starbucks is incorrect, you're
going to see, you know, disruptions like this. This, unfortunately,
is just a way of life right now. As we said,
this is not unusual. We did not see the stock
price of Amazon move ye yesterday in any sort of
dramatic way despite all the dysfunctionality that people were seeing
(06:05):
on the ground, meaning investors were not spooked by this,
perhaps thinking that this is part of just the normal
course of business. It is frustrating, no doubt, but it
is not unusual. I do think the big takeaway here
has to come with their market power, right We talked
about this a little bit before the break. Amazon accounts
for about thirty seven percent of the cloud computing market.
Another third goes to just two companies, Microsoft and Google.
(06:28):
We've centralized a lot of power in the hands of
just a few companies, and when something goes wrong, it's
going to have big ripple effects, an outsized effect. When
a flaw crops up like this DNS problem did yesterday,
do they.
Speaker 2 (06:41):
Know what caused it? Was it just a glitch in
the system. Yeah or okay, it was.
Speaker 3 (06:48):
Not a cyber attack, I think is the important thing
to say. Amazon came out very quickly, you know, at
the beginning of yesterday and said, this is a tech issue.
This isn't you know, any nefarious actors or foreign governments
or anything like that. This is just something went wrong, frankly,
and we don't know the specific nature of what went wrong,
but we know it has to do with their domain
name service. This is again a technology that translates you know,
(07:12):
human language KFI dot com, abcnews dot com for example,
into something that a computer can understand, into an IP
address or an Internet protocol basically a big string of
numbers that tells the computer where to go on the internet,
where to direct your request. That is what went wrong.
And then obviously, you know, you break one thing in
the morning and it actually ends up breaking like four
or five other things. So you saw this kind of
(07:33):
play out over the course of the day. And in fact,
right now we're actually seeing, according to down Detector, some
things related to Amazon Web services start to report more problems.
I think we're just going to be dealing with this
for the next you know, few hours. At least the
next few days, perhaps as Amazon continues to figure out,
you know, the long tail effect of this large break
(07:54):
yesterday morning.
Speaker 1 (07:55):
All right, So let's say I'm a business and I
rely on AWS to keep my business going, and I
can prove that this outage cost me money.
Speaker 2 (08:05):
Do I have any chance?
Speaker 1 (08:07):
Do I have any mechanism of making a claim and saying, hey,
I want my three hundred dollars, I want my ten
thousand dollars that you cost me.
Speaker 3 (08:15):
You know, that's a really good question. We have not
really heard anything on that front.
Speaker 1 (08:20):
You know.
Speaker 3 (08:20):
I think that a point against that type of class
action lawsuit or that legal action would be the fact
that like this happens to everyone, and this happens somewhat frequently.
Speaker 2 (08:29):
Right.
Speaker 3 (08:29):
Amazon Web Services itself went down in a very similar
way back in twenty twenty three, and in twenty twenty one,
and in twenty twenty. Microsoft and Google have dealt with this.
We talked about crowd Strike earlier, which is that cybersecurity
outfit that shipped a faulty bit of code. You know,
this is going to happen to some degree or another.
I think the lesson that those businesses can take here
(08:50):
has more to do with diversification. Right, instead of putting
all your eggs, your all your cloud computing into you know, AWS,
maybe you put like two eggs into that bad and
then two eggs into the Google basket, and then maybe
another one into you know, a smaller How do.
Speaker 2 (09:05):
You do that? How do you do that with one business?
Speaker 3 (09:08):
It's very easy. I mean, let's let's go back to
like the cafe example that we were talking about earlier.
You know, you do different things online with your small business.
Some is payments processing, some as payroll, some as you know,
menus or ordering or whatever it happens to be. Well, okay,
one feature can be hosted on this service, another future
can be hosted over here. Does that increase your costs? Well,
(09:28):
you gotta have to You're gonna have to answer that
for yourself. But like that's a way to guard against,
you know, your entire business going offline if one of
these major companies sees a disruption.
Speaker 1 (09:39):
Fair enough, Mike, thank you. That was a terrific way
of explaining this to us. We'll talk again next week.
Speaker 2 (09:46):
Sounds good care, have a good one. ABC News reporter
Mike Dubusky, Boy, he knows this stuff. Doesn't he okay
It's Tech Tuesday with Mike Dubuski, ABC News reporter and
tech maven out of New York.
Speaker 2 (00:07):
Mike, good morning, Thank you for joining us.
Speaker 3 (00:09):
Yeah, good morning.
Speaker 2 (00:10):
How are you all right?
Speaker 1 (00:11):
The big news is the AWS meltdown, and it is
absolutely enormous.
Speaker 2 (00:18):
And so here's my first question. Because they are known.
Speaker 1 (00:23):
I think they're the world's biggest company that does cloud computing.
Speaker 2 (00:28):
They are.
Speaker 1 (00:29):
And you know, I'm very confused because I thought your
information in the clouds were in the clouds and that's
not how it works, is it. And if you could
explain how this actually works and then what happened and
what's going on right now?
Speaker 2 (00:45):
Yeah.
Speaker 3 (00:45):
Sure. So if you're a small business, let's say you
want to open a cafe in downtown Los Angeles or
what have you, it doesn't really make economic sense for
you to spin up like a computer server to run
your payroll or to run your payments processing or whatever
else it is you do online. It kind of makes
more economic sense for a lot of companies to outsource
(01:06):
that work to someone who knows what they're doing, you know,
a Microsoft or a Google, or in this case, an Amazon,
which accounts for the largest chunk of the cloud computing service.
This is a service that works, you know, ninety nine
percent of the time. The reason that Amazon accounts for
about thirty seven percent of this market and Google and
Microsoft account for another third of it goes back to
(01:28):
the idea that they run really robust systems that don't
go down ninety nine percent of the time. Well, yesterday,
in the case of Amazon, we saw that one percent
of the time crop up when an issue with specifically
their DNS cropped up. This is a domain name service. Basically,
to make a very technical thing kind of simple, it
is a piece of technology that translates human language into
(01:50):
something that a computer can understand. It's what you type
into a web browser being translated into an IP address.
So that is the sort of nature of the problem.
We don't know specifically what went wrong with it, but
that's what went down, and it created these sort of
knock on effects throughout the day. Because Amazon Web Services
is so big and it hosts so many different you know,
servers and technologies out there, you break one thing, it
(02:12):
actually ends up breaking like four or five things. So
there was like a kind of knock on effect, a
whack a mole situation going on where Amazon engineers fix
the main problem pretty quickly and then you know, realize
these later problems down the line. The good news is
that most things are resolved at this point. A quick
look at down detector right now finds that most of
the major players who are experiencing issues yesterday among them
(02:34):
then most Snapchat, Microsoft three sixty five, Google It excuse
me a zoom, I should say not Google, but like,
these are all systems that are now running relatively normally,
and you know, the people are somewhat back to normal.
You're a Starbucks order on the app should work this morning.
Speaker 1 (02:50):
Now, when you say thirty percent of the market that
Amazon has, is that just the American market?
Speaker 2 (02:56):
Are we talking internationally?
Speaker 3 (02:58):
It's the global market?
Speaker 2 (02:59):
And to it.
Speaker 3 (03:00):
Yeah, I was talking to a friend of mine who
is a tech journalist in the UK, and they actually
experienced this in a much more acute way because they're
ahead of us. This outage actually happened for you guys
kind of in the overnight hours. For us here on
the East coast, it was just after three am, and
most of America is asleep at that point, so the
sort of initial wave of problems were only noticed by
people who were up at that time. Of course, we
(03:22):
started to see more reports trickle in, you know, nine o'clock,
ten o'clock, as people go to work and figure out,
you know, hey, this thing I had access to I
don't have access to anymore. Well, in the UK, that
hit their government systems, right, the website that controls their
equivalent of the IRS went down in part because of this,
So you know, it really does underscore exactly how much
influence Amazon has in this particular market. We saw similar
(03:45):
stories come out of continental Europe and beyond in Australia
as well.
Speaker 2 (03:49):
I mean, it is astounding how big this is. Now.
Speaker 1 (03:54):
If I'm not mistaken, this AWS accounts for more money
and more prime than the retail end of Amazon itself.
Speaker 2 (04:02):
Do I have that right? It does. Yeah.
Speaker 3 (04:04):
We often think of Amazon as like this e commerce website,
that's where they make their money. Of course, that's the
public facing component of their business. But when you actually
look at their profits, the majority of them come from AWS,
come from this cloud computing service. Amazon is a cloud
computing company before it's an e commerce retailer. That's the
long and short of it. And you know that again
(04:25):
is something I don't think a lot of people understand it.
Speaker 2 (04:27):
It's really important episode.
Speaker 1 (04:28):
Now, this cloud servicing, it has to do with these
massive forms of computers that keep my information, your information's
process things. I'm almost thinking that the world of computing
is growing far much quicker than their ability than people
(04:53):
like the WS or companies to deal with it and
the amount of energy that it takes.
Speaker 2 (04:58):
You want to talk about that.
Speaker 3 (05:00):
Yeah, So I think that there are certainly protocols that
companies like Amazon can implement to make sure that they're
not shipping out faulty code that breaks things like we
saw yesterday, or maybe they raise the standards in the
coding for you know, how they deal with third parties
and what have you. We also, in recent years have
seen generative artificial intelligence play a bigger role in sort
(05:20):
of like entry level coding, in fact taking the place
of entry level coders at many tech companies. However, humans
are imperfect, right, They're going to ship out improper code
every once in a while, and so is generative artificial intelligence.
Despite the fact that we talk about it all the
time as this transformative technology, it's still pretty bad at
a lot of things and hallucinates and makes mistakes frequently.
(05:43):
And that means that in the same way that you
come across a typo in the newspaper every once in
a while, or your order at Starbucks is incorrect, you're
going to see, you know, disruptions like this. This, unfortunately,
is just a way of life right now. As we said,
this is not unusual. We did not see the stock
price of Amazon move ye yesterday in any sort of
dramatic way despite all the dysfunctionality that people were seeing
(06:05):
on the ground, meaning investors were not spooked by this,
perhaps thinking that this is part of just the normal
course of business. It is frustrating, no doubt, but it
is not unusual. I do think the big takeaway here
has to come with their market power, right We talked
about this a little bit before the break. Amazon accounts
for about thirty seven percent of the cloud computing market.
Another third goes to just two companies, Microsoft and Google.
(06:28):
We've centralized a lot of power in the hands of
just a few companies, and when something goes wrong, it's
going to have big ripple effects, an outsized effect. When
a flaw crops up like this DNS problem did yesterday,
do they.
Speaker 2 (06:41):
Know what caused it? Was it just a glitch in
the system. Yeah or okay, it was.
Speaker 3 (06:48):
Not a cyber attack, I think is the important thing
to say. Amazon came out very quickly, you know, at
the beginning of yesterday and said, this is a tech issue.
This isn't you know, any nefarious actors or foreign governments
or anything like that. This is just something went wrong, frankly,
and we don't know the specific nature of what went wrong,
but we know it has to do with their domain
name service. This is again a technology that translates you know,
(07:12):
human language KFI dot com, abcnews dot com for example,
into something that a computer can understand, into an IP
address or an Internet protocol basically a big string of
numbers that tells the computer where to go on the internet,
where to direct your request. That is what went wrong.
And then obviously, you know, you break one thing in
the morning and it actually ends up breaking like four
or five other things. So you saw this kind of
(07:33):
play out over the course of the day. And in fact,
right now we're actually seeing, according to down Detector, some
things related to Amazon Web services start to report more problems.
I think we're just going to be dealing with this
for the next you know, few hours. At least the
next few days, perhaps as Amazon continues to figure out,
you know, the long tail effect of this large break
(07:54):
yesterday morning.
Speaker 1 (07:55):
All right, So let's say I'm a business and I
rely on AWS to keep my business going, and I
can prove that this outage cost me money.
Speaker 2 (08:05):
Do I have any chance?
Speaker 1 (08:07):
Do I have any mechanism of making a claim and saying, hey,
I want my three hundred dollars, I want my ten
thousand dollars that you cost me.
Speaker 3 (08:15):
You know, that's a really good question. We have not
really heard anything on that front.
Speaker 1 (08:20):
You know.
Speaker 3 (08:20):
I think that a point against that type of class
action lawsuit or that legal action would be the fact
that like this happens to everyone, and this happens somewhat frequently.
Speaker 2 (08:29):
Right.
Speaker 3 (08:29):
Amazon Web Services itself went down in a very similar
way back in twenty twenty three, and in twenty twenty one,
and in twenty twenty. Microsoft and Google have dealt with this.
We talked about crowd Strike earlier, which is that cybersecurity
outfit that shipped a faulty bit of code. You know,
this is going to happen to some degree or another.
I think the lesson that those businesses can take here
(08:50):
has more to do with diversification. Right, instead of putting
all your eggs, your all your cloud computing into you know, AWS,
maybe you put like two eggs into that bad and
then two eggs into the Google basket, and then maybe
another one into you know, a smaller How do.
Speaker 2 (09:05):
You do that? How do you do that with one business?
Speaker 3 (09:08):
It's very easy. I mean, let's let's go back to
like the cafe example that we were talking about earlier.
You know, you do different things online with your small business.
Some is payments processing, some as payroll, some as you know,
menus or ordering or whatever it happens to be. Well, okay,
one feature can be hosted on this service, another future
can be hosted over here. Does that increase your costs? Well,
(09:28):
you gotta have to You're gonna have to answer that
for yourself. But like that's a way to guard against,
you know, your entire business going offline if one of
these major companies sees a disruption.
Speaker 1 (09:39):
Fair enough, Mike, thank you. That was a terrific way
of explaining this to us. We'll talk again next week.
Speaker 2 (09:46):
Sounds good care, have a good one. ABC News reporter
Mike Dubusky, Boy, he knows this stuff. Doesn't he okay
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
CrimeLess: Hillbilly Heist
It’s 1996 in rural North Carolina, and an oddball crew makes history when they pull off America’s third largest cash heist. But it’s all downhill from there. Join host Johnny Knoxville as he unspools a wild and woolly tale about a group of regular ‘ol folks who risked it all for a chance at a better life. CrimeLess: Hillbilly Heist answers the question: what would you do with 17.3 million dollars? The answer includes diamond rings, mansions, velvet Elvis paintings, plus a run for the border, murder-for-hire-plots, and FBI busts.