https://www.rt.com/usa/536576-facebook-offices-chaos-dns-outage/
Facebook plunged into chaos as employees ‘CAN’T ACCESS’ servers – or even their workspaces
Facebook’s offices are as much of a mess for employees as its social media platforms are for users, according to anonymous workers who’ve spoken to reporters in the midst of the apparent companywide breakdown.
As the major Facebook outage dragged into hours, a Reddit poster claiming to be a member of the “recovery/investigation team” wrote the companywide snafu was indeed a “global outage for all FB-related services/infra[structure]” and explained that while repairs were underway, “the people with physical access is [sic] separate from the people with knowledge of how to actually authenticate to the systems and people who know what to actually do, so there is now a logistical challenge with getting all that knowledge unified.”
Facebook’s DNS [Domain Name Servers] were down due to what several commenters suggested was a configuration change that took effect right before the mega-crash.
“Something BIG seems to have died inside Facebook’s network,” Jim Salter, a reporter at ArsTechnica tracking the crashing-and-burning at Facebook HQ, observed, while DDoS protection provider Cloudflare’s senior vice president reported the entire Facebook network was unreachable.
A New York Times tech reporter had earlier confirmed that internal tools and communications platforms were down, including employee email and Workplace, which employees from different divisions use to communicate. “No one can do any work,” employees told him, likening the chaos to a “snow day” at the trillion-dollar company.
Reportedly, employees can’t even get into some of the buildings they need to access to evaluate the extent of the outage, as worker badges aren’t opening doors like they are supposed to.
Rumors and theories about what happened at Facebook have proliferated, including a report that the personal data of some 1.5 billion users is being sold on a dark-web hacker forum, coinciding with the outage. However, the alleged dump is said to have initially been posted two weeks ago, and may not be related to the Monday events.