The Facebook 2014 Outage: Why 'War Rooms' Are Bad for Deep Investigation
2025-02-23
The author recounts the epic Facebook outage of August 1st, 2014, dubbed "Call the Cops." Working in a cramped, overheated 'war room', the author found it impossible to effectively troubleshoot the root cause. He ultimately retreated to his own comfortable workspace. After 18 days of investigation, he pinpointed the problem: a process called 'fbagent' incorrectly sent a termination signal to all processes, leading to system failure. This experience highlights the importance of providing a suitable personal work environment during emergencies and emphasizes the value of in-depth investigation over rapid fixes.