World news and discussion on TalkFuse.
Microsoft’s massive outage offers a lesson in how to bring the world back online
Microsoft 365’s email service suffered a cascading outage on 12 November, but restoring it proved harder than finding a fix. The incident shows why recovery planning matters: as users reconnect, demand can overwhelm the systems that have just come back online.
Microsoft said no email was lost and there was no attack; at the worst point, about 80% of affected servers were unhealthy. Engineers restored capacity region by region, bringing each area fully online before users returned, after an earlier restart failed under renewed demand. Other outages that week affected services at the FAA, Workday, JPMorgan Chase, the NSA and Apple, though the article says there was no evidence they were connected.
In short
- Microsoft 365 email went down in a cascading server failure.
- About 80% of affected servers were unhealthy at the worst point.
- Engineers restored service region by region to manage demand.
Explained simply
Microsoft 365 is a collection of online workplace tools, including Outlook email, used by businesses and public organisations. These services run on large networks of computer servers, so a problem affecting part of the network can disrupt access for many customers.
When an online service is restored, users often try to reconnect at once, creating a surge in demand. If the recovering systems cannot handle it, they may fail again, so companies plan how to restore services gradually and manage demand.
Such outages can interrupt work across organisations that rely on online services. Several other major services also had outages that week, but no link between those incidents was established.