OK Google, who is the main ISP in Africa and how did he manage to direct your traffic to China via Russia? • The Registry

The prolonged explosion of Google's cloud and websites has been triggered by an unpleasant network update by a West African phone, it says.

One of the ISP operators in Lagos, Nigeria, operating a submarine cable between Portugal and South Africa, said that a faulty configuration at the end caused the redirection of Google traffic to China Telecom for 74 minutes.

At that time, web browsers and apps that tried to connect to Google, YouTube, or Google Cloud sites and platforms such as Spotify and Nest were routed to the Chinese phone via TransTelekom ISP in Russia and fallen into a black hole.

Gafa was possible because Main One leaked the details of a part of the Internet aspect in the configuration of another, temporarily reloading the spinal cord. Packages moving to Google have been sent on routes that have taken them around the globe.

Primary One colleagues with Google by agreeing to exchange traffic between them through a turning point. Simply, ISP has accidentally left out details of its routes on the Google Network in a way that has prompted the rest of the network to adjust their paths so that Google's traffic is headed for China Telecom.

"MainOne has a peering relationship with Google via IXPN in Lagos and has direct routes to Google that has leaked to China Telecom," Ameet Naik's cloud monitoring biz ThousandEyes explained today. These leak routes have spread from China Telecom through TransTelecom to NTT and other ISPs. We also noticed that this leak was primarily driven by business transit providers and did not affect both consumer ISPs. "

This eventually caused a bunch of Internet traffic buzzers to Google in the US and possibly elsewhere to sink into a bottomless ditch in China's Telcom network, effectively hitting the offline giant in the eyes of many netizens. It is understood that the data were not intercepted or handed over during the spill. Here's the main thing:

I researched the advertising @google prefixes one of our upstream partners. This was an error during a planned network upgrade due to a wrong encounter on our BGP filters. The error was corrected within 74 minutes and the processes introduced to avoid recurrence

– MainOne (@Mainoneservice) November 13, 2018

Google also said that none of its servers or data was affected by the incident.

"We are aware that some of the Internet traffic has been affected by incorrect IP routing and access to some Google services has been affected," said Chocolate Factory spokesman He Reg. "The root cause of the problem was Google's external and there was no compromise on Google services."

If not, disclosure will diminish fears that the interruption was the result of an attack or other unfavorable activity. At the same time, realizing that a thing as simple as a faulty regional ISP to configure a server could trigger a global interruption does not happen well either.

"This incident further highlights one of the fundamental weaknesses of the internet structure," Naik said. "BGP was designed to be a chain of trust between well-trained Internet service providers and universities that believe they have the information they receive. It has not evolved to reflect the complex and geopolitical trade relations between ISPs and nations today [Route Origin Authorization] there are few ISPs I use. Even corporations like Google, with massive resources available, are not immune to this type of BGP leakage or malicious hijacks. "

As noted by NSA adviser and former White House cyber security chief Rob Joyce, the incident should serve as a call to reassess the status of the BGP system.

"I hope this latest fiasco to re-orientate trafficking through China is the wake-up call for all who are serious about addressing the massive and unacceptable vulnerability inherent to today's BGP routing architecture," said Joyce said Tuesday. ®