Cloudflare outage broke 20% of the entire internet, which led to $500M-$2B loss in just a few hours

[BigBallsLarry]

Cloudflare went down today, for around 4 hours, and in just those few hours it's speculated that over $500M in revenue was lost globally.

If you were wondering why .org, .com, .is and other forums were offline - that's the reason.

But it wasn't just the forums, as cloudflare is responsible for the protection of over 20% of sites on the entire internet - and ALL of them went down during the outage (including twitter, spotfy, even downdetector and worst of all OpenAI)

Cloudflare Outage November 18 2025: How a Single Infrastructure Failure Took Down ChatGPT, X, and 20% of the Internet - Axis Intelligence

Breaking: Cloudflare global outage on Nov 18, 2025 took down ChatGPT, X/Twitter, and 20% of the internet. Technical analysis, timeline, and impact assessment. Cloudflare Outage November 18 2025

axis-intelligence.com

---

The reason it happened was pretty intresting, according to the article linked above, today (Nov 18, 2025), cloudflare pushed out a config file that was entirely auto-generated

I originally thought it was AI, but i was mistaken. Cloudflare operates on a human-written script - It collects new data (IP blocks, traffic rules, etc.), builds a config on it, and pushes it out to the system for better security.

So what triggered an outage of over 20% of sites globally - was one singular rogue file that was too large for it own system. It tried too much data and basically couldn't hold it's self together.

That's basically the gist of it, but you can read more about it here:

https://www.reuters.com/business/elon-musks-x-down-thousands-us-users-downdetector-shows-2025-11-18

And if you're wondering just how 20% of sites can have an impact of even BILLIONS of dollars, cloudflare is responsible for:

- CDN's

What is a content delivery network (CDN)? | How do CDNs work?

A content delivery network is a distributed group of servers that caches content near end users. Learn how CDNs improve load times and reduce costs.

www.cloudflare.com

- Security
- Routing
- API layers
- Enterprise services
- And handling/hosting millions of sites

And the biggest offender of it - OpenAI, which is the biggest AI company as of the moment, restricted user access for the entire duration of the outage. Leading to millions in loss.

So when it crashes, the entire digital economy comes to a full halt - therefore completely shitting itself.

It really shows how badly centralized the internet nowadays is, because if one of the main puzzle pieces goes rogue, half of the internet goes with it. A single oversized file caused MILLIONS of sites to go down, led to global financial damage and disrupted millions of users world-wide.

This is the fourth global-scale outage that happened in half a year, along with META, AWS, Microsoft and now Cloudflare. We can only hope nothing keeps crashing anymore.

@Jason Voorhees @Daddy's Home @Luca_. @Randomized Shame @TechnoBoss

 

[Asoka]

Is that why org is always down or is that’s just orgs fault

 

[Psocho]

interesting, thanks for the info

 

[BigBallsLarry]

Is that why org is always down or is that’s just orgs fault

That's just today, usually it's just .org servers themselves not keeping up.

 

[BigBallsLarry]

interesting, thanks for the info

Glad you found it intresting, i also did.

 

[Jason Voorhees]

Lesson 1738 of my life. never push a config larger than your median RAM per process without staged rollout + size gating. Canary ring died in 11 seconds, full fleet in 94 seconds. Classic Devops disaster.

 

[ChrisFletcher]

This is the fourth global-scale outage that happened in half a year, along with META, AWS, Microsoft and now Cloudflare. We can only hope nothing keeps crashing anymore.

@Jason Voorhees @Daddy's Home @Luca_. @Randomized Shame @TechnoBoss

I cant believe everyone in tech is getting layoff raped would lead to things in tech not functioning properly

 

[Starborn]

Cloudflare went down today, for around 4 hours, and in just those few hours it's speculated that over $500M in revenue was lost globally.

If you were wondering why .org, .com, .is and other forums were offline - that's the reason.

But it wasn't just the forums, as cloudflare is responsible for the protection of over 20% of sites on the entire internet - and ALL of them went down during the outage (including twitter, spotfy, even downdetector and worst of all OpenAI)

Cloudflare Outage November 18 2025: How a Single Infrastructure Failure Took Down ChatGPT, X, and 20% of the Internet - Axis Intelligence

Breaking: Cloudflare global outage on Nov 18, 2025 took down ChatGPT, X/Twitter, and 20% of the internet. Technical analysis, timeline, and impact assessment. Cloudflare Outage November 18 2025

axis-intelligence.com

---

The reason it happened was pretty intresting, according to the article linked above, today (Nov 18, 2025), cloudflare pushed out a config file that was entirely auto-generated

I originally thought it was AI, but i was mistaken. Cloudflare operates on a human-written script - It collects new data (IP blocks, traffic rules, etc.), builds a config on it, and pushes it out to the system for better security.

So what triggered an outage of over 20% of sites globally - was one singular rogue file that was too large for it own system. It tried too much data and basically couldn't hold it's self together.

That's basically the gist of it, but you can read more about it here:

https://www.reuters.com/business/elon-musks-x-down-thousands-us-users-downdetector-shows-2025-11-18

And if you're wondering just how 20% of sites can have an impact of even BILLIONS of dollars, cloudflare is responsible for:

- CDN's

What is a content delivery network (CDN)? | How do CDNs work?

A content delivery network is a distributed group of servers that caches content near end users. Learn how CDNs improve load times and reduce costs.

www.cloudflare.com

- Security
- Routing
- API layers
- Enterprise services
- And handling/hosting millions of sites

And the biggest offender of it - OpenAI, which is the biggest AI company as of the moment, restricted user access for the entire duration of the outage. Leading to millions in loss.

So when it crashes, the entire digital economy comes to a full halt - therefore completely shitting itself.

It really shows how badly centralized the internet nowadays is, because if one of the main puzzle pieces goes rogue, half of the internet goes with it. A single oversized file caused MILLIONS of sites to go down, led to global financial damage and disrupted millions of users world-wide.

This is the fourth global-scale outage that happened in half a year, along with META, AWS, Microsoft and now Cloudflare. We can only hope nothing keeps crashing anymore.

@Jason Voorhees @Daddy's Home @Luca_. @Randomized Shame @TechnoBoss

shit scared me I litterally commented on mastersprofile and then he viewed mine and then the site went down i deadass thought he fucking ip banned me

 

[ChrisFletcher]

Lesson 1738 of my life. never push a config larger than your median RAM per process without staged rollout + size gating. Canary ring died in 11 seconds, full fleet in 94 seconds. Classic Devops disaster.

Have you tried vibepushing it

 

[BigBallsLarry]

I cant believe everyone in tech is getting layoff raped and things in tech arent functioning properly

I know right? what a shocker?

 

[59H390]

Dang both org and that went down in the same day

Shi u said its cause of this that org was down

How come?

 

[Luca_.]

I’ve read somewhere that cloudfare has a lava lamp room, which is recorded by a camera, which is further turned into code. I don’t know the full story but from my understanding, the ,,password” is created from images of lava lamps.

 

[BigBallsLarry]

Lesson 1738 of my life. never push a config larger than your median RAM per process without staged rollout + size gating. Canary ring died in 11 seconds, full fleet in 94 seconds. Classic Devops disaster.

It's intresting how one file can cause all of this, I'm convinced the internet is held together by duct tape plus hopes and dreams.

 

[BigBallsLarry]

Dang both org and that went down in the same day

Shi u said its cause of this that org was down

How come?

It's simple, org uses cloudflare, so when cloudflare goes down so does org.

 

[ChrisFletcher]

I’ve read somewhere that cloudfare has a lava lamp room, which is recorded by a camera, which is further turned into code. I don’t know the full story but from my understanding, the ,,password” is created from images of lava lamps.

He saw this in an instagram reel btw
I know because i also saw the samw reel

 

[ChrisFletcher]

It's simple, org uses cloudflare, so when cloudflare goes down so does org.

All of my autist foruns went down at the same tkme

 

[unkownincel]

Interesting, I'll read it soon

 

[BigBallsLarry]

I’ve read somewhere that cloudfare has a lava lamp room, which is recorded by a camera, which is further turned into code. I don’t know the full story but from my understanding, the ,,password” is created from images of lava lamps.

I've seen that aswell, it's because lava lamps are completely unpredictable and random and change patterns permanently, so it's impossible to hack through.

 

[BigBallsLarry]

All of my autist foruns went down at the same tkme

They all use cloudflare

 

[Jason Voorhees]

It's intresting how one file can cause all of this, I'm convinced the internet is held together by duct tape plus hopes and dreams.

Gobal economy runs on a distributed system that can be bricked by a few lines of code. It proves how non-technical the tech world is at the very top, yeah.The engineers who actually touch the metal know exactly how fragile this house of cards is

 

[BigBallsLarry]

Gobal economy runs on a distributed system that can be bricked by a few lines of code. It proves how non-technical the tech world is at the very top, yeah.The engineers who actually touch the metal know exactly how fragile this house of cards is

It's crazy to me, do you think it'll improve with time (especially with ai) or will the developers be too scared to touch something set in stone?

I know games and sites that still use faulty code and assets from the 2000-2010 era of the internet, just because they're scared of upgrading.

 

[ChrisFletcher]

If you were wondering why .org, .com were offline

Holocaust for socially inept indian teens gays and trannies

 

[Daddy's Home]

I knew that during the day, but I didn't read the reason why it happened, so thanks Larry for that..

Also, the fact that Cloudflare alone has such a large monopoly over all these things is crazy as fuck, and the fact that they didn't already have the necessary measures in place to prevent what happened today is even stranger..

As you said, things like this just show how fragile this whole magical thing in front of our screens that we call the internet really is..

 

[Jason Voorhees]

It's crazy to me, do you think it'll improve with time (especially with ai) or will the developers be too scared to touch something set in stone?

I know games and sites that still use faulty code and assets from the 2000-2010 era of the internet, just because they're scared of upgrading.

Not really. Forget games even many goverment websites and airport security runs on 1970s software. US social security runs on 1970s COBOL, airport security managed by windows xp. nobody dares touch code that still works because the blast radius of a bad upgrade is measured in billions of dollars and human lives. Al won't fix it it'll just let us generate even more unreadable config at 100× speed, then merge it because AI says it is fine.

 

[BigBallsLarry]

I knew that during the day, but I didn't read the reason why it happened, so thanks Larry for that..

Also, the fact that Cloudflare alone has such a large monopoly over all these things is crazy as fuck, and the fact that they didn't already have the necessary measures in place to prevent what happened today is even stranger..

As you said, things like this just show how fragile this whole magical thing in front of our screens that we call the internet really is..

The internet is just an entire jenga game, and cloudflare, AWS, google cloud and vastly are the bottom blocks.

 

[EmperorVon]

Old news

 

[mewcel420]

I’ve read somewhere that cloudfare has a lava lamp room, which is recorded by a camera, which is further turned into code. I don’t know the full story but from my understanding, the ,,password” is created from images of lava lamps.

cloudflare cannot be real at this point bro

 

[BigBallsLarry]

Not really. Forget games even many goverment websites and airport security runs on 1970s software. US social security runs on 1970s COBOL, airport security managed by windows xp. nobody dares touch code that still works because the blast radius of a bad upgrade is measured in billions of dollars and human lives. Al won't fix it it'll just let us generate even more unreadable config at 100× speed, then merge it because AI says it is fine.

I figure this'll lead to many issues in the future though.. when those systems become less and less operable with time, and everyone is forced to upgrade leading to rushed results.

Maybe it's just my speculation. But i doubt this is even close to the end of global outages.

 

[Swancel]

I thought it was the WEF doing some testing for the future internet censorship. I didn't realize how many websites I use on a daily basis rely on cloudfare until today.

 

[BigBallsLarry]

Old news

org reliably came back only 2 hours ago

 

[BigBallsLarry]

I thought it was the WEF doing some testing for the future internet censorship. I didn't realize how many websites I use on a daily basis rely on cloudfare until today.

Me neither, i guess you never know the significance of something untill you lose it.

 

[Swancel]

Me neither, i guess you never know the significance of something untill you lose it.

I just exercised & listened to Fuentes's latest stream during the outage

 

[Water Bomb]

Was wondering what the issue was today.

 

[BigBallsLarry]

I just exercised & listened to Fuentes's latest stream during the outage

Based, i took a nap

 

[takethewhitepill]

im so thankful tbh. was not going to sleep even though my exam is today otherwise

 

[zinsa]

high iq but you joined this year summer and already 13k posts have you gone outside in the past few months?

 

[Swancel]

Based, i took a nap

I always find plenty of things to do.

 

[BigBallsLarry]

high iq but you joined this year summer and already 13k posts have you gone outside in the past few months?

Yes, i go outside plenty and still find time to post here as a hobby in the pass time

 

[Orka]

Cloudflare went down today, for around 4 hours, and in just those few hours it's speculated that over $500M in revenue was lost globally.

If you were wondering why .org, .com, .is and other forums were offline - that's the reason.

But it wasn't just the forums, as cloudflare is responsible for the protection of over 20% of sites on the entire internet - and ALL of them went down during the outage (including twitter, spotfy, even downdetector and worst of all OpenAI)

Cloudflare Outage November 18 2025: How a Single Infrastructure Failure Took Down ChatGPT, X, and 20% of the Internet - Axis Intelligence

Breaking: Cloudflare global outage on Nov 18, 2025 took down ChatGPT, X/Twitter, and 20% of the internet. Technical analysis, timeline, and impact assessment. Cloudflare Outage November 18 2025

axis-intelligence.com

---

The reason it happened was pretty intresting, according to the article linked above, today (Nov 18, 2025), cloudflare pushed out a config file that was entirely auto-generated

I originally thought it was AI, but i was mistaken. Cloudflare operates on a human-written script - It collects new data (IP blocks, traffic rules, etc.), builds a config on it, and pushes it out to the system for better security.

So what triggered an outage of over 20% of sites globally - was one singular rogue file that was too large for it own system. It tried too much data and basically couldn't hold it's self together.

That's basically the gist of it, but you can read more about it here:

https://www.reuters.com/business/elon-musks-x-down-thousands-us-users-downdetector-shows-2025-11-18

And if you're wondering just how 20% of sites can have an impact of even BILLIONS of dollars, cloudflare is responsible for:

- CDN's

What is a content delivery network (CDN)? | How do CDNs work?

A content delivery network is a distributed group of servers that caches content near end users. Learn how CDNs improve load times and reduce costs.

www.cloudflare.com

- Security
- Routing
- API layers
- Enterprise services
- And handling/hosting millions of sites

And the biggest offender of it - OpenAI, which is the biggest AI company as of the moment, restricted user access for the entire duration of the outage. Leading to millions in loss.

So when it crashes, the entire digital economy comes to a full halt - therefore completely shitting itself.

It really shows how badly centralized the internet nowadays is, because if one of the main puzzle pieces goes rogue, half of the internet goes with it. A single oversized file caused MILLIONS of sites to go down, led to global financial damage and disrupted millions of users world-wide.

This is the fourth global-scale outage that happened in half a year, along with META, AWS, Microsoft and now Cloudflare. We can only hope nothing keeps crashing anymore.

@Jason Voorhees @Daddy's Home @Luca_. @Randomized Shame @TechnoBoss

500M-2B is a pretty large different

but yea still crazy

 

[BigBallsLarry]

500M-2B is a pretty large different

but yea still crazy

That's just the estimate, there's no confirmed number yet but it's definitely in that ballpark

 

[Daddy's Home]

The internet is just an entire jenga game, and cloudflare, AWS, google cloud and vastly are the bottom blocks.

Also, last month and this month, even as I write this, something exceptional is happening on the internet..

The world is changing, and it's changing very fast..

I'm thinking of starting a thread on this topic in the next few days, where I'll talk in detail about what's happening and what implications all this will have..

 

[BigBallsLarry]

Also, last month and this month, even as I write this, something exceptional is happening on the internet..

View attachment 4339678View attachment 4339682

View attachment 4339692View attachment 4339694

The world is changing, and it's changing very fast..

I'm thinking of starting a thread on this topic in the next few days, where I'll talk in detail about what's happening and what implications all this will have..

I'd be very interested to hear all about it

We will see just how much looksmaxing becoming popular will shape our world. I'm sure at first not many people will actually care, but maybe in a few years or a decade..

Also, i find it pretty intresting that this thread still gained a decent amount of views despite only 200 people being active

 

[BigBallsLarry]

@gorgeous @NumbThePain @Stacyslayerᛉ I'm sure you guys can find this thread intresting

 

[BigBallsLarry]

@Gengar

 

[Stacyslayerᛉ]

@gorgeous @NumbThePain @Stacyslayerᛉ I'm sure you guys can find this thread intresting

Holy fuck, OpenAI?
I wrote an Biology exam this morning, 0 knowledge, only AI (high T behavior) and everything went fine.

 

[BigBallsLarry]

Holy fuck, OpenAI?
I wrote an Biology exam this morning, 0 knowledge, only AI (high T behavior) and everything went fine.

It should've been fine in the morning if you're from europe. The outage happened between 1-4pm.

 

[DR. NICKGA]

Also, last month and this month, even as I write this, something exceptional is happening on the internet..

View attachment 4339678View attachment 4339682

View attachment 4339692View attachment 4339694

The world is changing, and it's changing very fast..

I'm thinking of starting a thread on this topic in the next few days, where I'll talk in detail about what's happening and what implications all this will have..

https://looksmax.org/threads/grey-shami-is-about-to-interview-giant-implants.1679797/post-23893163

Look at the rat utopia experiment, the last stage was looksmaxxed rats

 

[DxstZ]

Cloudflare went down today, for around 4 hours, and in just those few hours it's speculated that over $500M in revenue was lost globally.

If you were wondering why .org, .com, .is and other forums were offline - that's the reason.

But it wasn't just the forums, as cloudflare is responsible for the protection of over 20% of sites on the entire internet - and ALL of them went down during the outage (including twitter, spotfy, even downdetector and worst of all OpenAI)

Cloudflare Outage November 18 2025: How a Single Infrastructure Failure Took Down ChatGPT, X, and 20% of the Internet - Axis Intelligence

Breaking: Cloudflare global outage on Nov 18, 2025 took down ChatGPT, X/Twitter, and 20% of the internet. Technical analysis, timeline, and impact assessment. Cloudflare Outage November 18 2025

axis-intelligence.com

---

The reason it happened was pretty intresting, according to the article linked above, today (Nov 18, 2025), cloudflare pushed out a config file that was entirely auto-generated

I originally thought it was AI, but i was mistaken. Cloudflare operates on a human-written script - It collects new data (IP blocks, traffic rules, etc.), builds a config on it, and pushes it out to the system for better security.

So what triggered an outage of over 20% of sites globally - was one singular rogue file that was too large for it own system. It tried too much data and basically couldn't hold it's self together.

That's basically the gist of it, but you can read more about it here:

https://www.reuters.com/business/elon-musks-x-down-thousands-us-users-downdetector-shows-2025-11-18

And if you're wondering just how 20% of sites can have an impact of even BILLIONS of dollars, cloudflare is responsible for:

- CDN's

What is a content delivery network (CDN)? | How do CDNs work?

A content delivery network is a distributed group of servers that caches content near end users. Learn how CDNs improve load times and reduce costs.

www.cloudflare.com

- Security
- Routing
- API layers
- Enterprise services
- And handling/hosting millions of sites

And the biggest offender of it - OpenAI, which is the biggest AI company as of the moment, restricted user access for the entire duration of the outage. Leading to millions in loss.

So when it crashes, the entire digital economy comes to a full halt - therefore completely shitting itself.

It really shows how badly centralized the internet nowadays is, because if one of the main puzzle pieces goes rogue, half of the internet goes with it. A single oversized file caused MILLIONS of sites to go down, led to global financial damage and disrupted millions of users world-wide.

This is the fourth global-scale outage that happened in half a year, along with META, AWS, Microsoft and now Cloudflare. We can only hope nothing keeps crashing anymore.

@Jason Voorhees @Daddy's Home @Luca_. @Randomized Shame @TechnoBoss

uh oh

 

[gorgeous]

@gorgeous @NumbThePain @Stacyslayerᛉ I'm sure you guys can find this thread intresting

i do thank you