Cloudflare, a server that hosts millions of websites around the world, went down on Tuesday (2), hurting pages like Canva, Udemy and Discord, for example over 16 million internet applications and APIs use the system, according to the company website.
In a note released on Cloudflare Status, network problems started at 10:52 am (Braslia time) and are being investigated.
"Cloudflare is seeing network performance issues.
Customers may be experiencing 502 errors when accessing sites on Cloudflare.
We are working to mitigate the impact to Internet users in this region," the statement said.
Around 11:15 am, the company added a new message on its website, saying it fixed the issue see update at the end of the story.
A second statement further clarified that it was not an attack but a software implementation error.
Discord fell or does not open? Find out if gamers chat is down
The error that appears when trying to access the "502 Bad Gateway" websites, which means something like "bad gateway", in free translation.
When the user accesses a website, the browser sends a request to the server where the page is hosted.
In this type of failure, the server accepts the request but is unable to complete it.
Each error has a number that determines its origin.
Cloudflare goes down and users see message "502 Bad Gateway" Photo: Reproduction / dnetc
Want to buy a cell phone, TV and other discounted products? Meet the Compare dnetc
The Cloudflare Status site also reports locations that are affected by the instability.
In Brazil, the problem happens in Rio de Janeiro.
Other countries like India, Philippines, Serbia, Pakistan and Iraq also suffer from server error.
On Twitter, the company name reached second place in the worldwide Trending Topics.
Cloudflare crashed and took half the world's internet along UAHEUIHAUEIHAUIIDUUIHEAUIHEAUIH
– July 2, 2019
cloudflare destroyed my productivity today i swear chefin
– July 2, 2019
Cloudflare is an Internet performance and security company created in 2009.
In May 2019, the company announced that it provides services to "over 16 million domains, websites, APIs, mobile applications and anything connected to the Internet" beyond 175.
datacenters in over 75 countries, with 30Tbps network capacity.
UPDATE: At 11:15 am Cloudflare added a new status to its site, reporting that it fixed the issue. "Cloudflare has implemented a fix for this issue and is currently monitoring the results."
At 11:50 am, the company reported that the failure was global and lasted about half an hour.
"The large outage has affected all Cloudflare services globally.
We saw a huge CPU spike that caused primary and secondary systems to crash.
We closed the process that was causing the CPU spike.
Service restored to normal within 30 minutes.
Now we are investigating the root cause of what happened.
Cloudflare Blog Announcement:
"Today for about 30 minutes, visitors to Cloudflare websites saw" error 502 "caused by a large increase in CPU utilization on our network.
This CPU spike was caused by an incorrect software deployment that was rolled back.
reverted, the service returned to normal operation and all domains using Cloudflare returned to normal traffic levels, it was not an attack (as some have speculated) and we are very sorry that this incident occurred.
– complete death to understand how this happened and how to prevent it from happening again ".
Via Cloudflare (1) and Cloudflare (2)
I can't run Discord; what to do? Ask questions on the dnetc forum.
WhatsApp has stopped working? See possible causes and solutions