On 27 January 2015, Facebook and Instagram stopped working worldwide for about an hour. According to Facebook's own status page, both services were down from 10:10 pm to 11:10 pm Pacific time on 26 January, which was 06:10 to 07:10 GMT on the 27th, the morning in Europe, Africa and much of Asia. A hacking group hinted that it was responsible, and the hashtag #SocialMeltdown2015 spread on Twitter. Facebook later said the cause was a change it had made to its own configuration systems, and that it was "not the result of a third party attack."
Here is what happened that morning, why other apps went down too, why the hacking rumour spread, and what to do the next time a major platform goes dark.
What users saw
For most people, Facebook simply would not load. The website and the mobile apps on every platform failed, and many users got no explanation at all. Some saw Facebook's generic error message telling them that something had gone wrong and that engineers were working on it. Instagram, which Facebook had bought in 2012, failed at the same time, and an Instagram spokesperson acknowledged the outage on Twitter.
Because Facebook had hundreds of millions of daily users even then, the outage was noticed instantly. Complaints and jokes flooded Twitter, which stayed up, and the #SocialMeltdown2015 tag began trending as people tried to work out whether the problem was their phone, their internet connection or something bigger.
Which other services were affected
The dating app Tinder also had problems that morning. That was not a coincidence: Tinder relied on Facebook login to manage user accounts, so when Facebook's systems failed, many Tinder users could not sign in either.
Early reports also listed AOL Instant Messenger (AIM) and the workplace chat service HipChat. Those names came largely from a tweet posted by the hacking group Lizard Squad, which listed Facebook, Instagram, Tinder, AIM and HipChat under the tags #offline and #LizardSquad. Contemporary security reporting could not confirm that AIM and HipChat had actually been down, so their inclusion is best treated as unverified.
WhatsApp, which Facebook had bought in 2014, appears to have kept working. Reports from that morning did not show it going down with the others, although technology reporters had been unsure at first whether it would.
The hacking rumour and Lizard Squad
Lizard Squad was a group that had become notorious weeks earlier for knocking the PlayStation Network and Xbox Live offline over Christmas 2014 with denial-of-service attacks, and for defacing the Malaysia Airlines website the day before the Facebook outage. Its tweet about Facebook appeared to take credit and was retweeted thousands of times.
That made the idea of a cyberattack seem plausible, and some early news coverage described the outage as part of a wider attack. But there was no technical evidence of one. A group claiming credit after the fact costs nothing, and hacking groups have often done exactly that to gain attention during outages they did not cause.
Within hours Facebook gave a clear explanation. A spokesperson said the problem began after engineers introduced a change that affected the company's configuration systems, that it had moved quickly to fix it, and that both Facebook and Instagram were back to full operation. Facebook also told reporters it did not expect any further break in service.
Why a small change can take down a giant
It can seem strange that one of the world's biggest websites could be knocked out by its own engineers. In practice, this is one of the most common causes of large outages. Big platforms push configuration changes, settings that control how thousands of servers route traffic, handle logins and talk to each other, many times a day. If a bad setting spreads quickly to every server, the whole system can fail at once, and recovering may take time because the tools used to fix the problem depend on the same systems.
The same pattern returned on a much larger scale on 4 October 2021, when Facebook, Instagram, WhatsApp and Messenger went down for around six hours. That outage was also caused by a faulty change, this time to the network routers that connect Facebook's data centres to the internet, and not by an attack. The 2015 incident was an early reminder of how much of the internet rests on a few companies' internal systems.
How to tell an outage from a hack
When a platform suddenly goes dark, rumours spread faster than facts. A few habits help:
- Check official status pages first. Most large services, including Meta's, publish status dashboards for developers and businesses.
- Look for confirmation from the company or from reliable technology news sites before believing a claim of responsibility posted on social media.
- Treat claims of credit with caution. A group announcing that it caused an outage is not evidence that it did.
- Remember that time zones matter. An outage late at night in California is a morning outage in London, which is why the 2015 event felt like a "morning" crash to much of the world.
Protecting your account during an outage
Outages are a favourite moment for scammers, who send messages that claim to fix the problem or warn that your account has been hacked. While a service is down:
- do not click links promising to "restore" your account, and never type your password into a page you reached from a message;
- do not rush to reset your password or delete the app, which rarely helps when the fault is on the company's side;
- turn on two-factor authentication once the service is back, so that your account is protected if your password ever leaks;
- keep another way to reach friends, family or customers, such as email or a different messaging app, especially if your business depends on one platform.
That last point is not theoretical. When a single company controls the page through which you reach your audience, an outage, or a sudden removal, can cut you off without warning. Thinking Humanity learned that first hand when Meta unpublished our Facebook page with 7.2 million followers. For a very different story about losing access to a social account, read about an influencer's deleted Instagram account and her public appeal.
Sources: HotHardware, 27 January 2015: Self-inflicted wound causes hour-long Facebook, Instagram outage; WeLiveSecurity (ESET), 27 January 2015: Facebook, Instagram and Tinder hit by outages, but deny cyberattack; The Register, 27 January 2015: Facebook, Instagram, Tinder outage; originally reported by Softpedia.
COMMENTS