Just over a week after a massive outage at Amazon Web Services (AWS) took down a range of websites and cloud-based services, Microsoft’s Azure cloud platform was hit by a service disruption that affected its services and customer applications, including the communications protocol that runs Dynamics 365 Contact Center.
Between 15:45 UTC on October 29 and 00:05 UTC on October 30, configuration change caused node failures in Azure Front Door (AFD), Microsoft's global, cloud-based service that handles content delivery and routing, according to Azure’s status page.
A whole host of Azure services experienced latencies, timeouts and errors, including Azure Communication Services (ACS). The voice and SMS messaging channels in Dynamics 365 Contact Center are built on top of ACS, which enables organizations to provision phone numbers, enable voice calls and manage SMS capabilities through ACS directly.
The outage disrupted customer services across multiple industries, including airlines, retail and gaming, all of which use Azure for their backend operations including point-of-sale systems, customer portals, and mobile apps.
Alaska and Hawaiian Airlines, which have recently merged, reported that the Azure hosts several of their key services, including their websites, which meant that customers were unable to book or check in for flights online.
Travelers who were unable to check in were advised to see an airport agent for their boarding pass and plan for additional time in the terminal, as there were long lines.
“Our teams worked quickly to stand up our backup infrastructure to allow our guests to book and check-in for their flights online while minimizing operational disruptions,” Alaska Airlines stated. The airline added that it would bring the affected systems back online "once Microsoft resolves the issue from its end.”
Heathrow Airport also reported problems with its website, while retailers like Starbucks, Costco, and Kroger experienced interruptions to their websites and mobile apps, indicated by spikes in reports on outage monitoring site DownDetector.
Microsoft’s own services were not immune, as its customers were unable to log in to Xbox Live, Minecraft, and other Microsoft gaming platforms, and access to 365 productivity tools like Outlook, Teams and SharePoint went offline for enterprise users as well as consumers.
The timing of the outage was ironic, given that Microsoft reported its quarterly earnings with its investor relations website down ahead of the analyst call.
The company reported 40 percent year-on-year growth in its Azure revenues, indicating how deeply entrenched the cloud infrastructure is across industries.
The outage served as a reminder of the critical role cloud infrastructure plays in modern customer service delivery and the risks of centralized dependencies.
