Facebook, Instagram and WhatsApp taken down for 14 hours by suspected BGP leak from European ISP
Longest downtime in Facebook's history caused by BGP leak, according to Netscout
Facebook, Instagram and WhatsApp were taken down for 14 hours yesterday in a global outage caused by a Border Gateway Protocol (BGP) leak from a European ISP, according to network management specialist NetScout.
The downtime was the longest in Facebook's 15-year history, with the company issuing a denial addressing widespread rumours that it had been targeted and taken down by a distributed denial-of-service attack (DDoS).
The services went down at around 4pm in the afternoon in the UK and were only brought back in the early hours of Thursday morning. During the outage users were still able to view Facebook but couldn't post anything.
"While not malicious in nature, such events can prove disruptive on a widespread basis. It is very important that all network operators implement BGP peering best current practices," said NetScout principal engineer Roland Dobbins.
However, network monitoring company ThousandEyes told Computing's sister site The Inquirer that the cause of the outage appears to be internal to Facebook, rather than a network or internet delivery issue.
"We saw '500 internal server errors' from Facebook. Given the sheer scale and continuous changes that these web scale providers are constantly making to their applications and infrastructure, sometimes things break as a result of these changes, even in the most capable hands," the company claimed in a statement.
It continued: "We're not seeing any BGP changes that are affecting connectivity, packet loss or latency. Since Facebook uses its own backbone network, it's not clear [and] we don't have insight as to how an external transit route issue would cause a disruption within the internal Facebook network."
For Facebook, the outage could have a significant impact on revenues, with the company losing tens of millions of dollars in advertising revenues.
"The internet celebrated its 30th anniversary this week. The whole objective was that the internet was a self-healing web. ie: one bit goes down but the system mends itself and keeps functioning. But if users just rely a few sites we all become very vulnerable," commented TechMarketView's Richard Holway.
He continued: "At least Facebook going down just meant fewer cat videos were posted. But if WeChat went down in China many people's everyday lives would be affected as they couldn't make payments or order stuff too. A bit like Facebook, Google, Amazon and all the online banking systems going down at the same time."
BGP has been at the centre of a number of internet traffic routing incidents in recent years, with claims that authorities in China and Russia have hijacked traffic several times for reasons as-yet unknown.
Further reading
More on Cloud and Infrastructure
How to secure yourself against Malware during the COVID-19 pandemic
Security is always one of CIOs' top headaches, and the global pandemic hasn't improved matters. So what do organisations need to do in order to better protect themselves during the COVID-19 crisis?
HashiCorp moves into cloud-ops managed services
HCP is analogous to MongoDB Atlas in that it will provide a cloud-based service offering ‘push-button' access to the company's products via a single control plane, CTO says
Huawei may be a risk but the West should compete with Chinese firms rather than banning, says ex-Google CEO Eric Schmidt
'The Chinese are just as good, and maybe better, in key areas of research and innovation' Schmidt says
CIA's specialised hacking unit failed to protect its own systems from hackers, investigation finds
CIA discovered the breach only after classified information was published by WikiLeaks
