If you've had or are having problems using websites and apps today, it might well be due to the Microsoft Azure outage.
While the IT breakdown isn't nearly as bad as the CrowdStrike boot-loop crisis from earlier this month (not that that was exactly Microsoft's fault) various big names are experiencing connectivity problems they blame on Azure's network infrastructure multi-hour meltdown.
In other words, those brands are relying on Microsoft's cloud to power their applications and websites, and when those remote services fall over, as they've done today, they bring down those customers.
Customers may have experienced issues connecting to Microsoft services globally
The Windows giant said the issue lies with its network infrastructure, its Azure Content Delivery Network, and related CDN-like Azure Front Door service, which together means "customers may have experienced issues connecting to Microsoft services globally."
It all started to fall apart around 1145 UTC today, the day Redmond is due to reported its latest quarterly financial results.
"An unexpected usage spike resulted in Azure Front Door (AFD) and Azure Content Delivery Network (CDN) components performing below acceptable thresholds, leading to intermittent errors, timeout, and latency spikes," Microsoft noted at 1900 UTC.
"We have implemented network configuration changes and have performed failovers to provide alternate network paths for relief. Our monitoring telemetry shows improvement in service availability from approximately 1410 UTC onward."
Interestingly, while the service is being fixed, the cure may cause a little more disruption. Microsoft explained:
We continue to investigate and mitigate reports of specific services and regions that are still experiencing intermittent errors. While our previous network configuration changes successfully mitigated the impacts of the usage spike, these changes caused side effects for a small number of services.
Our updated mitigation approach has successfully reduced these impacts across regions in Asia Pacific and Europe.
Our phased approach is now applying this same mitigation targeting regions in the Americas, and we expect that services should be fully mitigated by 2030 UTC.
So in about an hour's time, from publication, Redmond reckons all should be well again. Let's take a look at the impact of Azure's instability.
Unsurprisingly, Minecraft maker Mojang, a subsidiary of Microsoft, says users may "may experience some intermittent failures connecting to our services due to issues with Azure." While Minecraft has both single and multiplayer modes, it does require users to authenticate their account before downloading and playing the game, and if these services fail then no more playtime.
Incidentally, this vulture tried to see what would happen if he tried to install the Windows version of Minecraft on his PC, and discovered that the launcher would spit out an error message and also prevented the Java version from being launched, likely because the Microsoft Store is also facing downtime. So that's fun.
Microsoft 365 is another Redmond-operated service confirmed to be impacted, and although there have been some murmurings about Xbox Live going down too, so far the official Xbox status page says everything is fine. Let us know in the comments section if that matches your experience.
Game creation engine Unity additionally reported problems with Azure that is preventing downloads and uploads. MongoDB and Moesif also said the Azure outage caused them headaches, respectively for their Cloud and web portal services.
Microsoft-owned GitHub said its Codespaces tool had issues around 0930 EST, and about half an hour later DocuSign noticed problems with its Rooms feature. Both diagnosed the problem as having originated from an outage at third-party, which we assume is Azure, tho it's not confirmed.
GitHub and DocuSign claimed the issue was fixed around the same time, which may imply Microsoft is indeed making progress on bringing Azure back up.
Starbucks also reportedly experienced an outage with its mobile app, though the coffee corporation hasn't responded to The Register's inquiries about whether this is related to Azure. But we suspect it might not be a coincidence since outage report statistics for Starbucks on Downdetector are very similar to Microsoft 365's numbers. ®
Source: The register