Microsoft’s public-cloud service Windows Azure experienced a widespread outage at the end of February due to a software bug related to a leap-year time miscalculation, the company’s Server and Cloud corporate VP Bill Laing wrote in a blog post.

“While final root cause analysis is in progress, this issue appears to be due to a time calculation that was incorrect for the leap year,” he wrote.

The Azure operations team first became aware of the issue on 28^th February at 5:45 pm PST. The team deployed a fix to most of the cloud’s sub regions and restored availability to most customers by about 3 am the next morning.

At that time, some sub-regions and customers were still experiencing isues.

The disruption affected the compute portion of the cloud portfolio, which includes a variety of other services, such as storage, networking, as well as Service Bus and SQL Azure.

While storage and SQL Azure services were not impacted, services dependent on compute did experience disruptions. They included access control service, service bus, SQL Azure Portal and data sync services.

Laing apologized for the incident’s negative effects on Microsoft’s customers and said the company would proactively issue a service credit to all those impacted.