Facebook, Inc., the largest and most popular social network, went down on Monday for the second time this week and third time in two weeks. Neither the app nor the website were functioning for an undetermined number of users.
The website started to load very slowly and finally crashed after showing a specific message: “Sorry, something went wrong. We’re working on it and we’ll get it fixed as soon as we can.”
Facebook’s platform status dashboard noted there was a “major outage” beginning at about 12 p.m. PT. The dashboard monitors the performance of the Facebook Graph API (application program interface), which is used by many developers that integrate their services with Facebook.
Soon after the dashboard confirmed the issue, it also went down and was inaccessible. “A Facebook-wide issue is causing the Facebook Graph API to be temporarily unavailable. We’re working with our core infrastructure teams to identify the issue and will update you when we have more information,” the message read.
This event counts for the third time the service has gone down in three weeks. The first time it was on September 16 and second, only four days ago.
On September 23, Facebook also went down for approximately 2.5 hours. According to a post wrote by Robert Jhonson, this was the worst outage they have ever had in over four years.
Johnson explained that the reason was “unfortunate handling of an error condition”. Apparently, the automated system that verifies configuration values of the page did not work very well with invalid configurations in the cache.
The system that attempted to correct configuration values ended up being turned off and Johnson promised to look for new designs for this configuration system.