“DNS problems affect several Microsoft 365 and Azure services,” explains Microsoft on Twitter and refers to its own status page, on which there should be more information about the failures. However, this is not the case (anymore). Instead, it says there are no problems. Nevertheless, on Friday night at least Teams, Xbox Live and SharePoint were partially unavailable.
The history can also be followed on Twitter. The first problems appear there late on Thursday evening. “We are in the process of investigating issues that are apparently causing users to no longer have access to their Microsoft 365 services. More information will follow.” The information does not go any further than the subsequent explanation that there were failures in the DNS system. About three hours later, the all-clear that all services are available again.
Again and again failures of the cloud services
Microsoft only just had similar problems: a good two weeks ago, several services failed due to an authentication error. Specifically, the cause of this would be an error in a key in the Azure Active Directory (AAD), which was automatically deleted although it was still needed. And last fall, Microsoft’s cloud services were down worldwide due to an AAD change. At that time, you couldn’t register for five hours. This failure was also during the night for Central European time, so it is unlikely to have affected many companies and other users in this country. But it was different with previous problems.
At Google, too, there have been multiple disruptions in the cloud services, which made numerous other services unavailable; including mails, calendars and collaboration tools. At the same time, Google offers premium support for companies called Mission Critical Services (MCS), which is aimed at customers who cannot afford service failures – for example in the financial sector. The service should not only guarantee failure-free operation of the cloud, but also develop concepts and methods for the operation of fail-safe applications with the company.