Mark Zuckerberg’s Facebook, Instagram and other social media platforms crashed last night for the second consecutive month. Several banks, phone networks, and other tech giants also suffered major outages in recent months.
Even Britain’s biggest supermarket Tesco was brought to its knees by a hack of its website and app last month, leaving thousands of customers unable to order groceries for 48 hours and costing the retailer an estimated £40m in lost revenue.
What’s the real reason behind these website outages or crashes? Is this just a coincidence or a problem with the back-end systems? Or is something more sinister happening?
MailOnline spoke to a variety of internet security and cyber experts to determine the root causes of the outages. We started with Meta, the parent company behind Facebook, Instagram, and WhatsApp.

Technical problems: Facebook and Instagram crashed last night for the second consecutive month, while banks and other businesses also suffered outages in recent weeks
Matthew Hodgson is co-founder and CEO at Element, and technical co-founder at Matrix. He said that Meta’s centralised backend system was a major problem.
This means that Facebook, Instagram, WhatsApp, and Messenger can all be affected by a single failure, instead of just one.
‘The spate of recent outages is an inevitable side-effect of massive centralisation, where companies like Facebook have ended up on the critical path of providing infrastructure for billions of people,’ Mr Hodgson told MailOnline.
“Consumers end-up unwittingly being obliged to put all of their eggs in one basket and when inevitably a failure mode for that company or its infrastructure occurs (be it accidental, malicious), the end result can be catastrophic.”
Professor Bill Buchanan, an Internet scientist, also believes that the internet is too centralised.
He calls for multiple nodes to ensure that a service doesn’t fail on its own.
Hodgson also agreed.
‘The solution is to decentralise apps like Facebook and WhatsApp, just as the web and email and internet itself has no central points of control or failure — so there’s simply no single company or infrastructure which can have an outage which impacts the whole system,’ he said.
Jake Moore, a spokesman for internet security and antivirus company ESET, told MailOnline: ‘Centralising their data has been one of the biggest issues for Meta combining all three giants — Facebook, WhatsApp and Instagram.
“This brings problems that are often not discovered until the crunch time, which can be too late.
“These outages are often the result and companies using these platforms to conduct business must have additional tools in place should it rely on these services, such as another messaging platform.
“We are likely to experience more outages in coming months because more people use these services.”
Many experts believe that an outage is caused by a cyber-attack.
But they added that more often than not it’s down to human error, as was the case last month when Facebook, Instagram, WhatsApp and Messenger went down for seven hours.
It was ultimately attributed at the bottom to a faulty upgrade that disconnected Meta’s servers and brought down all its platforms.
A single unnamed IT customer was also responsible for a massive internet blackout that brought down hundreds of websites around the globe in June.
Millions of people were unable to access major sites like Amazon, Spotify, PayPal, and the UK government.
The outage was caused by a software bug triggered when a customer for Fastly — the US cloud-computing company responsible for the problems — changed their settings, the firm said.

Meta, which is responsible for Facebook, Instagram, and WhatsApp, is headed by Mark Zuckerberg.
Gav Winter, CEO at RapidSpike.com and cybersecurity firm RapidSpike.com said that large-scale outages have increased over the past 12 month. He also stated that human error was often a major culprit.
He stated that mistakes were made because employees are under pressure from their companies and choose to take unwise shortcuts.
MailOnline was also told by experts that outdated systems that keep the internet connected are making IT problems worse.
The World Wide Web was created in 1989 and is now an ‘aging infrastructure’ that is under increasing pressure from users.
Professor Buchanan previously stated to the BBC that the internet is not the large-scale distributed network DARPA (the Defense Advanced Research Projects Agency), tried to create. It can withstand a nuclear strike on any part of it.
“The protocols it uses are essentially the ones that were created when we connected to mainframe computers via dumb terminals. A single flaw in its core infrastructure can bring it all to the ground.

MailOnline spoke with a number cyber security experts in order to understand why there have been so many outages. The first was the Meta issues.
Moore stated that tech giants and other businesses were also affected by the unexpected surge in traffic caused by the Covid pandemic. This strains their infrastructure.
He added that these outages are increasing in number due to the sheer volume of online users and traffic.
“The pandemic is forcing more people online in rapid succession than the slow increase predicted over time.
“Sheer large numbers of people are funnelling through passages made for times past cause blackouts, much like when many people visit a website at the same time to purchase tickets that are just on sale.
Moore stated that businesses should test their infrastructure, and have multiple failsafes. But as often happens with the case it is simply impossible to simulate its size and magnitude in a safe environment.
According to Luke Deryckx, chief technical officer at Down Detector, a website that monitors websites for disruption, widespread outages are becoming more frequent and more serious.
MailOnline spoke to experts who confirmed that major outages have been on the increase and are only going to get worse.
They believe that the answer lies in companies shifting to more decentralised systems, updating old infrastructure, and creating servers that can host more users.
There will be more outages until then, with Meta and its large user base bearing the brunt.