Microsoft resolved a networking issue impacting Azure, Outlook, Teams, and other services globally for hours; services in China and for governments were not hit
This closes out the same outage Reuters tracked a day earlier, when Teams, Azure and Outlook went dark across India, Japan, the UK and other markets for hours. It is also the latest entry in a long series: Microsoft has previously blamed a DNS traffic spike for a similar multi-service failure (Teams, Azure and Office 365 hit in 2021) and an account authentication problem before that.
The notable detail in today's resolution is what did NOT fail: services in China and government deployments stayed up, which points to how isolated or sovereign configurations absorbed a fault that propagated through the globally shared network layer.
First-order effects
Enterprise customers running daily work on Teams, Azure and Outlook lost hours of productivity worldwide, and Microsoft now owes those customers a root-cause explanation and any SLA credits its contracts allow.
Second-order effects
Rivals selling workspace and cloud alternatives gain a fresh reliability talking point, pushing enterprise buyers to weigh multi-provider or multi-region setups rather than defaulting to a single Microsoft stack.
Third-order effects
The recurrence — authentication in 2017, DNS in 2021, networking here, and later another Azure configuration error behind a Microsoft 365 outage — suggests single-provider concentration itself is becoming the risk enterprises and regulators price, favoring architectures where a shared network fault cannot take every service down at once.
The trend: As more business-critical work concentrates on one provider's shared infrastructure, individual network faults keep converting into economy-wide downtime, steadily raising the premium on redundancy and isolated deployments.
We've identified a potential networking issue and are reviewing telemetry to determine the next troubleshooting steps. You can find additional information on our status page at https://status.office365.com/ or on SHD under MO502273.
⚠️We are currently investigating a networking issue impacting connectivity to Azure for a subset of users. More information will be provided as it becomes available. For more information, please refer to https://azure.status.microsoft/ ...
🛠️Engineers have confirmed the issue impacting connectivity to Azure resources has been mitigated. A detailed resolution statement can be found in the Status History at https://status.azure.com/...
We've isolated the problem to networking configuration issues, and we're analyzing the best mitigation strategy to address these without causing additional impact. Refer to the admin center MO502273 or https://status.office365.com/ for more information.
At @Cloudflare we keep a careful eye on our ability to reach major clouds and hosting providers, and on error rates at them. Today's Azure outage shows up pretty clearly. https://twitter.com/...
Let's just agree that this is sh*t timing for Microsoft. Whether or not the layoffs has anything to do with outages directly, it's impossible to convince customers of that. The stress of not knowing if you or your team is impacted is in fact VERY real. https://www.bbc.com/...