A hidden software bug from July played a key role in Microsoft’s global outage. Millions of users faced problems with Outlook and other services in November. The company has now revealed how three separate failures combined.
Microsoft: A bug that stayed hidden for months
As reported by The Guardian, Microsoft linked the outage to a software flaw that had been present since July. The flaw affected servers that manage network traffic for Microsoft 365. Under certain conditions, it could use up server resources and stop machines from responding. The change had been rolled out across the world over several weeks. Yet it caused no clear trouble for more than three months. Microsoft’s tests failed to catch it.
The company used less powerful machines in its test labs. The bug was more likely to appear on newer servers with more processing cores. The problem came to light on November 12. It hit Outlook, Exchange Online and other Microsoft 365 services. Users in many countries could not access their email or other work tools. The disruption lasted for about 24 hours.
Three problems came together
The software flaw was only one part of the story. Microsoft said two other issues helped trigger the outage. Microsoft Teams had moved from an alternate network gateway to the main one about 15 hours before the disruption. This was a routine step. It meant that the main gateway had to handle more traffic. The same gateway also supported Exchange Online, which runs Outlook email.
At the same time, nearly 14 per cent of the gateway’s server capacity was already unavailable. This was due to a separate issue from the previous day. On its own, the loss of capacity had not caused a major problem. But the extra traffic put more pressure on the remaining servers. The three issues then combined during peak business hours in the United States.
Requests began to pile up. Servers became slow and stopped responding. Microsoft’s systems tried to move traffic to working servers. But this added more pressure to those machines. At the worst point, more than 80 per cent of the affected gateway servers were unresponsive.

Why recovery took so long for Microsoft?
Fixing the problem was not simple. When engineers brought servers back online, they faced a large queue of waiting requests. The traffic quickly overwhelmed them again. This slowed the recovery process. Microsoft also said its usual tools could not quickly remove the bug. The flaw had been introduced months earlier, so a simple rollback was not enough. Engineers had to prepare a targeted fix and return servers to service in stages.
The outage affected more than office work. Hospitals used paper records, schools struggled to contact parents, and some public services faced disruption. Microsoft apologised and said it would publish a fuller report. The incident is a reminder of how much daily work now depends on cloud services. A single fault may not always cause a major failure. But several small problems can combine and spread quickly. Microsoft will now need to show how it plans to prevent a repeat.
Author’s Opinion
The outage shows how one hidden flaw can affect millions when other systems fail at the same time. Microsoft has explained what went wrong, but users will now expect stronger tests and better backup systems. Trust depends on reliable service.
READ MORE: Meta fined €805m over Facebook accessing users’ photos.
ALSO READ: SpaceX makes history with Starship after booster loss.
READ NEXT: Microsoft restores Outlook and Teams after global outage.


Comments
0Be the first to join the discussion.