There was a major Microsoft Azure outage between 07:05 UTC and 09:45 UTC today (25 January 2023). Customers experienced issues with networking connectivity, manifesting as network latency and/or timeouts when attempting to connect to Azure resources in Public Azure regions.
This outage affected a majority of the 500 million users of the platform. However, direct users of Azure were not the only ones affected, because many other apps use Azure as a cloud dependency – and the users of those apps were also affected.
Let’s take a brief look at the Microsoft data, the details behind the outage from Metrist, and an overview of the other apps affected by the outage.
Microsoft’s Report
Azure updated their status page with the following information:
- The problem. According to Microsoft, the issue was “a change made to the Microsoft Wide Area Network (WAN) impacted connectivity between clients on the internet to Azure, connectivity between services within regions, as well as ExpressRoute connections.”
- How they fixed it. Once they identified the change, they rolled it back.
- Recovery times. Networking recovered at “09:00 UTC onwards across all regions and services with the “final networking equipment recovering at 09:35 UTC.” Further, “most impacted Microsoft services automatically recovered once network connectivity was restored.”
This is pretty high-level information, so let’s explore more details about the outage from Metrist functional testing.
Metrist Data
According to the Metrist functional tests and real-time status page data, we found the following.
- Outside-in Testing: Down. All Metrist monitors which had to connect to Azure services from outside Azure reported a Down during the outage – including most Azure DevOps Monitors (AzureCDN, Azure DevOps Artifacts, Azure DevOps Boards, Azure DevOps Repos, Azure Dev Ops Test Plans).
- Inside Testing: Not Impacted. Metrist was able to confirm that connectivity from within Azure itself was not impacted. Many Metrist checks run from within Azure itself and those checks were not impacted.
Red lines on the graphs below indicate functional tests that returned a “down” status.
Azure DevOps Artifacts
Azure DevOps Boards
Azure DevOps Repos
Azure DevOps Test Plans
AzureCDN
Other Affected Apps
Microsoft users were not the only ones affected by the outage. Many other apps have Azure as a cloud dependency, so users of these apps were also affected. Such software included
Github
Sentry
CircleCI
Eclipse Foundation Services
Netlify
And we could see this from their real-time status page feeds in the Metrist dashboard.
What You Can Do
Companies are relying more and more on Cloud Dependencies like Azure. And as we can see, outages can become widespread and affect many more people than the customers of Azure itself.
If you want to have real-time updates on outages and specific information about how the outages are affecting your software, sign up for Metrist.
