Microsoft's global outage ran into a second day, and the cause is still unconfirmed

44 Sources

Share

Microsoft's global 365 outage stretched into a second day, degrading Outlook, Teams and other workplace tools for users in at least 74 countries. The company ruled out a cyberattack but has not confirmed a cause.

Microsoft spent a second day on Friday recovering from a global outage that knocked out Outlook, Teams and other Microsoft 365 services for users in at least 74 countries and territories

8

6

4

9

10

. More than 430 million people use the workplace software suite, per Microsoft's own website, which makes any failure of it a worldwide event

4

10

.

The complaints piled up fast. Downdetector recorded roughly 130,000 reports of Outlook problems in the US on November 12, and reports arrived from 74 countries and territories within about an hour

4

5

7

9

. The disruption hit Outlook, Teams, Copilot, SharePoint and Exchange Online, and it landed unevenly

8

4

7

3

. Some users could receive email but not send it. Some Teams users messaged and met normally, and files stored directly in SharePoint stayed accessible

8

4

7

3

.

Source: New York Post

Source: New York Post

The explanation kept shifting

Microsoft first pointed to elevated load across multiple regions, then said, "We determined the elevated load isn't driven by an increase in traffic, but by a number of servers being unexpectedly unavailable; straining the remaining infrastructure"

6

. Later updates focused on servers failing automated health checks, with a core process on affected machines described as overwhelmed

6

2

4

.

Source: New York Post

Source: New York Post

The company ruled out a cyberattack, saying there was no indication of malicious activity, and no group claimed responsibility

2

4

7

1

9

10

. Microsoft did not answer TechCrunch's questions about how it made that check, such as whether it had logs capable of detecting malicious activity

1

. Per a report citing The Verge's Tom Warren, the outage is linked to changes in Azure network configuration; Microsoft has not confirmed it

3

.

The fixes struggled too. Engineers restarted affected servers, but the company said changes were not taking as expected and servers could become unresponsive again once traffic returned

6

2

10

. It paused software deployments and configuration changes while reviewing recent work, and found none that correlated with the start of the disruption

6

2

10

. Microsoft's own status channels struggled as well: the Microsoft 365 admin center was inaccessible at one point, and the fallback status page was down for at least one reporter

8

10

.

Governments, airlines and AI tools

Public services in at least seven countries reported disruption. Per the reports, UK government email went down, a National Health Service trial in Manchester was halted, police urged people to report crime by phone, and prescription delays were flagged

4

9

10

. Per the same reports, Singapore's government, Australia's domestic intelligence agency, Norway's ministries and Japanese government IT were affected

2

9

10

. So were the Philippines' Supreme Court and US local governments including Baltimore County, Jacksonville and Austin

2

9

10

.

Businesses felt it as well. Indian airlines IndiGo and Akasa Air reported customer-support email problems, Hong Kong banks saw partial disruption, and Workhorse Group delayed its quarterly earnings call

9

10

. Danny Lau, an aluminum firm owner, said he could not receive updates from US customers and feared production timelines could slip

9

10

. OpenAI and Anthropic both reported problems with tools connecting their AI services to Microsoft 365

2

4

9

10

. Troy Hunt, an Australian cyber-security expert, said Microsoft is "a massive part of the internet backbone" and that the outage reached everyday tasks like appointment reminders and account access

4

.

Recovery, and a familiar pattern

Progress came on Friday. Microsoft reported improving conditions across North America and much of Europe. It still warned of intermittent access issues and delayed email, and gave no deadline for full restoration

7

3

10

. It extended the email timeout to 48 hours so messages stuck in backlogs would not be lost

7

. The company said it was taking a measured approach to identify and repair the affected infrastructure, prioritizing the clearing of email queues and backlogs

10

. Doug Madory of network analyst firm Kentik said traffic was returning after about 18 hours of downtime

1

.

TechCrunch reported that Microsoft is likely to face questions over its slow recovery from a failure of its own making

1

. At least four significant Microsoft service disruptions have been documented since 2024: an Outlook and Teams outage in January 2026, an Exchange disruption caused by a code update that took days to fix in August 2026, a days-long cloud outage blamed on an authentication configuration issue in September 2026, and Azure connectivity problems in October 2025

2

1

9

10

. The July 2024 CrowdStrike-related outage grounded flights and hit hospitals, banks and retailers

9

10

.

Source: CRN

Source: CRN

© 2026 TheOutpost.AI All rights reserved