Twitter pins global outage on unscheduled 'system change'

The brief outage heavily disrupted several of the firm's APIs including Enterprise Search and its Engagement API

The Twitter logo on a card surrounded by other cards with images such as fingerprints and locks

Twitter is blaming a global outage that knocked its entire social media platform offline for several hours on Thursday evening on irregularities with some of its application programming interfaces (APIs).

A planned system change caused Twitter to be down for many users from approximately 10:30pm BST, with the full outage lasting an hour and a half, the company confirmed through its health status page. The incident was not caused by a breach or an external security threat.

Disruption, however, continued well into Friday morning, with an investigation finding several Twitter APIs, including Enterprise Search, Insights Track and the Engagement API, were also disrupted during the outage. Services continued to suffer from some degree of disruption for several hours, with services only recovering in full at 5:13am BST.

“Twitter has been down for many of you and we’re working to get it back up and running for everyone. We had some trouble with our internal systems and don’t have any evidence of a security breach or hack,” the company confirmed just after midnight.

“The outage was caused by a system change initiated earlier then planned [sic], affecting most of our servers. We’re working hard to bring Twitter back to normal and expect things to be fully resolved in 1-2 hours. We appreciate your patience.”

Related Resource

Reducing MTTR the right way

Best practices for fast incident resolution

Download now

There are no details as to what the nature of this inadvertent change to Twitter’s systems was. The incident, however, struck a day after the company was embroiled in something of a scandal surrounding the suppression of an article published in the New York Post about the son of US presidential candidate Joe Biden, Hunter.

Twitter had initially decided to block the article in part due to an internal policy of not sharing what might be construed as hacked material. Following backlash, however, the company changed course, according to the New York Times and decided to water down its policy and only block material shared directly by hackers, not third-parties such as newspapers.

Most Popular

The benefits of workload optimisation
Sponsored

The benefits of workload optimisation

16 Jul 2021
Samsung Galaxy S21 5G review: A rose-tinted experience
Mobile Phones

Samsung Galaxy S21 5G review: A rose-tinted experience

14 Jul 2021
Six ways boards can step up support for cyber security
Business strategy

Six ways boards can step up support for cyber security

22 Jul 2021