Billions of people and millions of businesses around the world have been partially cut off this Monday due to the fall of Facebook and all its services , among which is WhatsApp, Instagram and Facebook Messenger .
At 5:40 p.m. PDT, Facebook disappeared from the internet . The messages stopped being sent and received through WhatsApp, and the Facebook and Instagram applications showed errors when entering. The surprising thing about this fall, which has been total, is its duration. How can the services of one of the largest and most powerful companies in the world be inaccessible for six hours? This is the question that everyone has been asking. In the heads of users who don't understand the technical ins and outs, the internet cannot – or should – be as fragile as it really is.
Photo by Greg Bulla on Unsplash
from the fall of Facebook
As soon as the company's applications and services became inaccessible, Cloudflare found that its Domain Name System (DNS) stopped working alongside its IP address infrastructure . “It's as if someone had pulled the cable from their servers,” says the company on its blog.
DNS is a kind of Internet telephone book by which the names of web pages are translated to their real address, the IP address. It is used so that the user does not memorize a series of numbers instead of a simple name like facebook.com . When a user wants to access a website, DNS is responsible for locating the destination. They are in charge of translating an address that can be understood by humans into one that is understood by machines: (client-internet-server).
The problem was in a higher layer, since the service was untraceable even through IP addresses.
However, the problem was in a higher layer, since the service was untraceable even through the IP addresses . This resided in the Border Gateway Protocol (BGP). BGP is the key behind the autonomous systems that connect internet networks. Its job is to constantly list the route that can be taken. That is, DNS says where you want to go, BGP says where. The great internet routers that make it all work have constantly updated lists of possible routes to take. The BGP makes the network of networks possible, and allows the Facebook network to have a presence on the internet and therefore be accessible from anywhere in the world.
Facebook, due to an error in its configuration, announced new routes. These were inaccessible and, for practical purposes, the company disconnected itself from the internet. The problem could have been quickly solved by reverting the new settings, something that could not be done because the entire Facebook network was inoperative. The servers had to be physically accessed. But, as reported by several journalists, even the access to and from their offices depended on the network.
The solution? To enter into force
The office of one of the largest companies in the world spent hours without being able to use its internal communication network and without being able to access the computers on which the changes can be reversed. Until they managed to physically access their data center.
The first information indicated that those in charge of solving the incident had to use a rotaflex or angle grinder to access the room. This information has already been denied by the journalist Mike Isaac of the New York Times, that it has issued a rectification, although it is the testimony of employees speaking anonymously in front of the public discourse that Facebook may want. It is not convenient to acknowledge a disastrous and comical situation in the midst of the scrutiny the company has been subjected to in recent days. Officially, it has been recognized that there were difficulties due to “physical security”.
The company has issued the following statement hours after solving the problem, explaining the causes:
“The underlying cause of this outage also affected many of the internal tools and systems we use in our day-to-day operations, complicating our attempts to quickly diagnose and resolve the issue.
Our engineers have found that configuration changes to the routers that coordinate network traffic between our data centers caused problems that disrupted this communication. This disruption to network traffic had a cascading effect on the way our data centers communicate, causing our services to be disrupted. They are back online and we are actively working to get them back to normal.
We want to make it clear at this point that we believe the main cause of this outage was a faulty configuration change. We also have no evidence that user data was compromised as a result of this outage. ”
Photo by Brett Jordan
When the world depends on one company
Whether or not it is true that thousands of people have been able to communicate again or continue operating their businesses thanks to a rotaflex, the situation reopens an important debate: is it convenient that almost the entire internet and, therefore, society, turn into around a single company?
It is not only the obvious polarization problems that they promote and the harmful effects on the mental health of their users – according to multiple studies that are coming to light – but the fact that, gradually, Facebook has taken over what citizens see and consume on the internet and the world's largest communication channel, which is WhatsApp.
In turn, more and more businesses depend on it. Digital media that use their services as a gateway to their content, stores that use it as a showcase or the owner of a restaurant that manage orders through Messenger. The convenience and immediate benefits they offered in the past are now, even for six hours, lost millions of dollars and the feeling that much of our lives may depend on a few misplaced lines of code.
