Scores of websites and services went down Friday afternoon due to problems with Cloudflare’s DNS service, sparking rampant speculation about the cause. After all, a global DDOS attack would totally fit the real-life apocalypse movie that 2020 is increasingly turning into.
The outage, which started shortly after 5 p.m. ET, brought down popular sites and services like Discord, Politico, Feedly, and League of Legends for roughly half an hour on Friday. Once connections were restored, Cloudflare issued an incident report stating that the issue “was not as a result of an attack” and that it “has been identified and a fix is being implemented.”
Turns out the real explanation’s nothing so nefarious. Evidently, half the internet briefly went dark because of a crappy router in Atlanta.
“It appears that a router in Atlanta had an error that caused bad routes across our backbone. That resulted in misrouted traffic to PoPs that connect to our backbone,” Cloudflare CEO Matthew Prince tweeted Friday. “We isolated the Atlanta router and shut down our backbone, routing traffic across transit providers instead. There was some congestion that caused slow performance on some links as the logging caught up. Everything is restored now and we’re looking into the root cause.”
According to the incident report, this issue with Cloudflare’s 126.96.36.199 DNS service impacted its data centers internationally, from Frankfurt to Paris and Schiphol, as well as several in major U.S. cities, including Los Angeles, Chicago, Seattle, Atlanta, and San Jose. Reports on Downdetector showed the outages appeared to be concentrated in the U.S. and northern Europe.