Since 2004, Facebook Inc. has evolved into a colossal online platform. Not only does the company own Instagram and WhatsApp, but it also owns the virtual reality company ‘Oculus’, the digital wallet company ‘Novi’, the smart speaker company ‘Portal’, and ‘Workplace from Facebook’, a communication tool.
Description of the outage:
On Monday, October 4th, Facebook and its subset of applications—including Instagram, WhatsApp, and Oculus—were unavailable for hours, disrupting a critical communications platform used by billions. Users stated that the applications began flashing error messages at approximately 9:10 p.m. Indian Standard Time (IST). The outage lasted more than five hours, after which several applications progressively came back online. However, the company indicated that it would take time for the services to be restored.
Usually, a majority of these disruptions are fixed shortly. They are frequently localised, with some people unable to access a website that is accessible from a different country. This outage, on the other hand, was global, affecting all of Facebook’s various applications. As technicians addressed the situation, there were claims of “mayhem” across the Facebook headquarters. Later that day, the company stated that changes to Facebook’s core internet infrastructure, which handles traffic between its data centres, were the reason this time. This halted communications and spread to other data centres; the social media giant added, “bringing our services to a halt.”
The technical aspects:
Several internet traffic analysts and researchers have linked it to ‘Domain Name System’ (DNS) and ‘Border Gateway Protocol’ (BGP) problems. The internet is made up of autonomous systems, which are massive networks, and information is routed from these large networks to the rest of the internet via the BGP—a means for these autonomous systems to share routing data. In layman terms, DNS is the address system for each website’s location (its IP address), and Border Gateway Protocol (BGP) is the route that identifies the most efficient route to that Internet Protocol (IP) address. According to Cloudflare, Facebook essentially told BGP on Monday through a series of modifications that the pathways to Facebook no longer existed. As a result, users attempting to access Facebook and the other companies owned by it could not do so.
Facebook characterised the event in a blog post due to a faulty configuration update to its backbone routers. This had a cascading effect, and brought its services to a halt. Because all of Facebook’s internal systems are run from the same location, it was difficult for personnel to diagnose and fix the problem. The Guardian’s UK technology editor, Alex Hern, tweeted, “Facebook runs EVERYTHING through Facebook”—so, the traditional way you would resolve a problem of this kind was defunct. Facebook employees were reportedly unable to access their communication network and workplace due to the outage affecting the company’s security pass system.
How was it fixed?
According to those familiar with the situation, services were eventually restored when a team gained access to its server machines at a data centre in Santa Clara, California. The company expressed regret for the outage. Following the reinstatement of services, they wrote on Twitter––“To the huge community of people and businesses around the world who depend on us: we’re sorry. We’ve been working hard to restore access to our apps and services and are happy to report they are coming back online now. Thank you for bearing with us.”
Effect of the downtime:
When the outage began, users of Facebook and Instagram took to Twitter to express their vexation and make light of their inability to access these applications. The hashtag #facebookdown surfaced in the latest and trending section of Twitter, and the application itself saw a massive boost in digital activity. The incident spawned hundreds of new memes, and its suddenness emphasised the startling variability that underpins our burgeoning digital market.
According to Fortune, Facebook’s total income loss would be around $99.75 million. The figure is based on Facebook’s second-quarter earnings—which included $29.08 billion in revenue over 91 days. The calculation works out to $319.6 million every day—or $13.3 million per hour on average. The interruption had a more significant impact on Facebook CEO Mark Zuckerberg’s finances, whose worth fell by an estimated $6 billion after the Facebook stock fell 4.8 per cent in trade on October 4th.
While the Facebook and WhatsApp outages were amusing for some and only a minor inconvenience for others, it was a devastating interruption for many. In platforms that have become central to businesses, community life, and even government service administration for millions of people in developing countries, downtimes of this scale can be severely impactful. The incident was a reminder of Facebook’s hegemony over the modern communications market.
During the extended outage of Facebook Inc.’s network of applications and services, downloads and user sign-ups for Signal and Telegram, two private communication applications, skyrocketed. The official Twitter account posted, “Hello literally everyone,” eliciting comments from WhatsApp and Instagram and other accounts. According to The Verge, Facebook employees used Discord and FaceTime to communicate during the downtime.
According to CNBC, Facebook’s biggest outage in history occurred in 2008, when the social media platform was only four years old. It went dark for nearly a day due to a problem that didn’t allow many of the platform’s 80 million users to load their timelines. Facebook experienced a two-hour outage in 2010 due to a networking issue, and the company has also faced outages in 2013, 2014, 2015, and 2019.
There have been four major WhatsApp outages in 2020, the most recent in January and lasted roughly three hours. An outage occurred in April, followed by a two-hour interruption in July, and a brief outage in August. Furthermore, the recent disruption lasted far longer than a 2019 outage of Facebook-owned apps that lasted roughly an hour and was “was caused by technical problems,” according to WhatsApp.
Facebook’s response and other intensifying challenges for the company:
Facebook has stated that it is investigating what occurred on October 4th to “make our infrastructure more resilient.” There was “no evidence that user data was compromised,” according to the company.
The outage occurred at a particularly challenging time for the company, which has come under fire for its scope and social impact. Former Facebook employee Frances Haugen told CBS News that the corporation had put “growth over safety” as their objective. Before a US Senate committee, she said that the company’s websites and applications “hurt children, stoke division, and weaken our democracy.” She has amassed hundreds of pages of internal research, and has subsequently shared it with the mainstream media, lawmakers, and regulators, indicating that Facebook was aware of several adverse effects its services have.
Apart from its continuing economic impact, the unsolicited downtime gave us what some would call “a much-needed internet detox.” The outage demonstrated how deeply Facebook and its application ecosystem are ingrained in people’s daily lives—work-related and otherwise. Globally, they have a considerable lead amongst non-Facebook applications. The far-reaching ramifications of this temporary outage will provide further grist to those who press on the essentiality for more competitors in this domain, and push towards the reconceptualisation of internet sovereignty.
BBC, Forbes, The Guardian, The Verge, Fortune, CNBC
Written by Suhani Kabra for MTTN
Edited by Parva Mehrotra for MTTN
Featured Image via Pngtree