A costly outage of all Microsoft 365 services — including Outlook, Teams, SharePoint, Copilot and Defender — continued into Friday morning.
The software giant reported “encouraging progress” toward solving the crash, which began at 12:30 p.m. on Thursday. The outage has disrupted organizations globally and caused untold financial damage.
Microsoft officials said a recent network configuration change that occurred at the time of the outage may have overwhelmed a process that sets up secure connections on each server. The company said overnight that it was reverting the change, which caused servers to stop handling requests.
People were interested in these podcasts
According to the company, progress remained slow because servers returning to service “can become unresponsive again once they begin receiving traffic.”
Microsoft said users can expect delayed, not lost, email because the company extended its email timeout to 48 hours, and should switch to phone calls and texts for urgent matters while checking status updates.
Multiple media reports show that the outage has paralyzed or slowed courts, schools, airlines, brokerages, government offices and healthcare companies around the world.
Affected companies include Indianapolis pharmaceutical giant Eli Lilly and Co., which uses Microsoft 365 Enterprise for its roughly 40,000 employees.
The Michigan Department of Technology, Management & Budget and Detroit’s City Hall reported email problems. In Maryland, Baltimore County staff couldn’t send or receive email.
Agencies in other countries have also been affected – including Australian federal police and intelligence agencies and government systems in Japan and Singapore.
Among the more than 430 million people who use Microsoft 365, user-submitted reports of the outage to Downdetector peaked at around 130,000 people – the same number of U.S. residents who use Outlook.
According to a CNN report, some courtrooms lost live feeds, and schools in the U.S. and Australia fell back on cellphones and text messaging.
NBC News reported that OpenAI and Anthropic, two leading tech giants relying on Microsoft connectors, were also affected.
No sources have estimated the cost of damages caused by the outage, as companies continue to pay idle staff on top of recovery and rework expenses.
For most large companies, an hour of downtime costs more than $300,000, according to a 2024 survey of more than 1,000 firms by the research firm ITIC, and 41% of respondents put the cost between $1 million and $5 million an hour.
A report in Dotcom-Monitor pointed to an estimate from insurer Parametrix that a July 2024 CrowdStrike outage caused an estimated $5.4 billion in direct losses at Fortune 500 companies, or roughly $44 million per affected company.
The actual cause of the Microsoft outage will remain unconfirmed until the company issues its customary post-incident report.
The Uptime Institute has estimated that power failure, network infrastructure faults and human error cause 39% of major outages.
Contact the author
Please read our comment policy before commenting.