Fastly: Global Site Crash Was Caused By a Software Bug


Fastly: The cloud server company and online service provider Fastly explained what happened last Tuesday morning (8), when several pages of global reach and from the most varied sectors were down for a moment.

According to the company, it all started with sending a software update to customers on May 12th. Without further details, this generated a bug that went unnoticed by the security team and was activated “by a specific customer configuration under specific circumstances”. As a result 85% of the network was affected by an error.

Platforms like Reddit, Spotify and Twitch, as well as news sites like CNN, BBC, The New York Times, The Guardian and The Verge were down for about an hour. By the end of the afternoon of the same day and with all services restored, the correction was already in progress.

what fastly says

“Even if specific conditions generated this drop, we should have anticipated it. We provide mission critical services and handle any action that may cause problems with the highest sensitivity and priority. sincerely to the community for their support,” says the statement.