A Microsoft Azure outage has impacted numerous web applications, with reports coming in on Downdetector from around 8:30 AM PDT yesterday. The outage, which has reportedly been fixed (according to Azure’s status page), affected the websites of London’s Heathrow Airport, Alaska Airlines, Costco, Office 365, Minecraft, and Xbox Live.
This Azure outage follows last week’s issues on Amazon’s cloud platform, which also affected many companies and apps.
What Caused the Microsoft Outage?
An “inadvertent configuration change” by Microsoft was the cause of this outage. The change initiated problems in Azure Front Door (AFD), their cloud content delivery service, which resulted in traffic routing issues and decreased availability of Azure for a range of services.
Microsoft’s Azure Front Door (AFD) service, crucial for directing internet traffic efficiently, experienced a significant disruption due to an internal update. This update inadvertently corrupted AFD’s routing configurations, causing Azure to lose the ability to properly direct user requests for a wide array of Microsoft and customer services.
As AFD is integrated with numerous Microsoft offerings, including Outlook, Teams, and Xbox, the issue rapidly escalated and affected many services.
Is Microsoft Still Down?
At the time of writing, Microsoft has confirmed that the Azure outage has now been “mitigated” – as reported on X by Azure Support.
“Engineers have confirmed that an issue which impacted a subset of Azure services is now mitigated.” Azure Support
In an official Azure statement, Microsoft explained that safeguards have “since been reviewed and additional validation and rollback controls have been immediately implemented to prevent similar issues in the future.
“Our team will be completing an internal retrospective to understand the incident in more detail and will share findings within 14 days,” the company wrote. “Once we complete our internal retrospective, generally within 14 days, we will publish a final Post Incident Review (PIR) to all impacted customers.”
Which Companies Were Affected?
According to The Independent, various reports of affected apps and websites rolled in over the hours-long outage, including Heathrow Airport, Alaska Airlines, Costco, Starbucks, NatWest, Vodafone, Minecraft, 365, Xbox, and more.
Alaska Airlines was seen on X responding to disgruntled customers who were facing long airport lines, website issues, check-in issues, and more.
How Long Did The Outage Last?
The outage lasted approximately eight hours, with customer impact being first felt at 10:45 AM PDT.
See the full list of Microsoft Azure health service updates below (all times PDT):
- 11:04 AM: Investigation commenced following monitoring alerts being triggered.
- 11:15 AM: Investigation begins to examine configuration changes within AFD.
- 11:18 AM: Initial communication posted to public status page.
- 11:20 AM: Targeted communications to impacted customers sent to Azure Service Health.
- 12:26 PM: Azure portal failed away from Azure Front Door.
- 12:30 PM: Microsoft blocks all new customer configuration changes to prevent further impact.
- 12:40 PM: Microsoft initiates the deployment of its ‘last known good’ configuration.
- 1:30 PM: Microsoft starts to push the fixed configuration globally.
- 1:45 PM: Manual recovery of nodes commenced while gradual routing of traffic to healthy nodes began after the fixed configuration was pushed globally.
- 6:15 PM: PowerApps mitigation of dependency, and customers confirm mitigation.
- 7:05 PM: AFD impact confirmed as mitigated for customers.
Summary
The recent Microsoft Azure outage, which lasted several hours, now appears to have been resolved. The companies affected – including Alaska Airlines, Heathrow Airport, and Office 365 – have all reported normal operating once again.
In total, the outage, as aforementioned, lasted approximately eight hours – nearly four hours longer than AWS’s the week before.
Have you been affected by the outage? Email tips@salesforceben.com
