Of the many news that happened yesterday perhaps the most commented (on Twitter, especially) was the worldwide downfall of Facebook, WhatsApp, and Instagram services. This article is an explanation of what the causes have been, or at least what we have been informed so far.
For security reasons it is likely that you will never know what exactly happened, therefore, the theory that your brother-in-law has explained that an intern hit a button down, is a possible explanation.
Unavailable page and worldwide downfall
On October 4, 2021, Facebook and its two acquired services such as WhatsApp and Instagram suffered a worldwide slump. It seemed that it was going to be a good day when suddenly when entering WhatsApp you appeared a message from Connecting… . In addition, facebook.com did not respond nor Instagram.
If that happens you do not have to turn off your mobile or computer, you just must check that you have an internet connection. You open a browser on your device and put a web page. If you can navigate your connection is fine, therefore, the only reasonable explanation is that WhatsApp is down or the service that does not respond as shown in Fig 1.
What matters is DNS
DNS (domain name server) is a huge database of IP addresses that, just as it works similar to a phone book, tells your browser or app where it has to go to get to a place on the Internet. A user types a web address, example domain.com, the browser will read the DNS list and looks for that address that has an IP address assigned, a server. And that server returns a response that shows the content as seen in Fig. 2.
In this sense, DNS 1.1.1.1 is one of the most used today and belongs to CloudFlare. In computing there is always a limit, in DNS too. It is a good idea to use a DNS server that is fast, since that way we save time waiting. In addition, this CloudFlare DNA has other advantages such as security, privacy, avoids blocks etc.
You can perfectly do nothing on your device and use the DNS provided by your internet service provider. You can also use Google’s 8.8.8.8 and 8.8. 4.4..
Therefore, if a page does not work many times, it is because the DNS are down or do not locate your page because they do not have it in their list. It is always the first thing that is checked when something fails. And the Internet if something has is that it is in a continuous state of error and solution, assembly and disassembly. It’s as if it was always breaking and they were repaired again.
Cloudflare shares what’s known about Facebook’s global downfall
Cloudflare is an Internet security and domain name server service used to protect your website against DDoS or similar attacks. It’s like a filter you put before your website to filter the content that reaches it.
In their forum they have explained the possible causes of what has happened. The crash began at 1651 UTC where they opened an internal incident because Facebook’s DNA search returned SERVFAIL, a server failure. Is it possible that Facebook is down? They reviewed the DNA solver 1.1.1.1, but realized that something more serious was happening.
Social media erupted, information its engineers quickly confirmed. Not only was Facebook down, but its WhatsApp and Instagram services were down as well. Their DNS names stopped resolving and their infrastructure IPs seemed to have disappeared. It was as if facebook’s servers had been a general blackout and they had disconnected from the network.
Meet BGP
BGP stands for Border Gateway Protocol. It is a mechanism for exchanging routing information between autonomous systems (AS) on the Internet. The big routers that make the internet work have huge, constantly updated lists of possible routes that can be used to deliver each network packet to its final destinations. That is, Facebook’s servers can change IPs but routers update their listing and re-channel a user’s request to a valid server.
Without BGP, Internet routers wouldn’t know what to do and the Internet wouldn’t work. The Internet is literally a network of networks, like a network of thousands of cables, and is linked by BGP. BGP allows a network (say Facebook) to advertise its presence to other networks that make up the Internet. What happened on October 4 is that Facebook did not advertise its presence, ISPs (Internet service providers) and other networks could not find the Facebook network and therefore it was not available.
Sin BGP, los enrutadores de Internet no sabrían qué hacer e Internet no funcionaría. Internet es literalmente una red de redes, como una red de miles de cables, y está unida por BGP. BGP permite que una red (digamos Facebook) anuncie su presencia a otras redes que forman Internet. Lo que ocurrió el 4 de octubre es que Facebook no anunciaba su presencia, los ISP (proveedor de servicios de Internet) y otras redes no podían encontrar la red de Facebook y, por lo tanto, no estaba disponible.
Meet ASN
Each of the networks has an ASN: an autonomous system number. An autonomous system (AS) is an individual network with a unified internal routing policy. An AS can originate prefixes (let’s say they control a pool of IP addresses), as well as transit prefixes (let’s say they know how to get to specific groups of IP addresses).
The PeeringDB website was created to facilitate interconnection between networks and interconnection coordinators. A database is no longer just for looking and related information. It now includes all types of interconnection data for networks, clouds, services and enterprises, as well as interconnection facilities that are being developed at the edge of the Internet. Facebook’s ASN is 63293 as can be seen in https://www.peeringdb.com/net/14490.
Each ASN needs to advertise its prefix paths to the Internet using BGP; otherwise, no one will know how to connect and where to find us. That is, your data must be sent to a database so that the rest of the routers know where they have to go.
Summary
The events that have occurred on October 4, 2021, with the worldwide downfall of Facebook, Instagram and WhatsApp, is a reminder of the antifragility of the Internet. Antifragility is a property of the Internet that means that not all things can be working perfectly all the time, but that the Internet is in a continuous state of assembly and disassembly. And this global downturn is a demonstration of this ownership of the Internet.
It’s a reminder that the internet is a very complex and interdependent system of millions of systems and protocols working together. All these systems allow 5 billion users to connect to the Internet every day and surf the Internet.
The downfall incident was resolved around 21:00 UTC, in which there was renewed BGP activity from the Facebook network that peaked at 21:17 UTC. Therefore, the Facebook network ceased to be available between 15:50 UTC until 21:20 UTC, where the DNS returned to work, and Facebook returned to the mounting state again.
