Redmond recently slotted in a (very) detailed apology explaining the Windows Azure Storage disruption that affected its Southern US region customers late December. The outage lasted for two days.
The disruption affected 1.8 percent of the company’s total Windows Azure Storage accounts. Microsoft says thousands of businesses may have experienced problems (with online services and websites) that started on December 28, 2012.
The company, however, once again flunked on providing a number. A Microsoft official let out in May last year that Windows Azure was powering “high tens of thousands” of customers.
An in-depth detail of the service outage was provided on the Windows Azure blog over at MSDN by Mike Neil, with elaborate explanation on the root cause of the interruption:
“On December 28th, 2012 there was a service interruption that affected 1.8% of the Windows Azure Storage accounts. The affected storage accounts were in one storage stamp (cluster) in the U.S. South region. We apologize for the disruption and any issues it caused affected customers.”
Microsoft attributed the outage to a human error, as described in the blog post.
The service was fully restored on December 30, 2012. Customers were, however, initially left in the dark about the outage for at least 1.5 hours. Windows Azure subscribers get updates about the service performance via Microsoft’s Primary Service Health Dashboard.
In this particular instance, though, the dashboard relied on the very storage stamp that had experienced problems. Storage stamps are what Microsoft calls its regional units that consist of multiple storage node stacks.
This isn’t the first time dashboard problems emerged in a Windows Azure service disruption. The dashboard also went down in February last year with a “leap year bug” service failure.
The company has, nevertheless, promised to improve service in the future, and also plans on crediting its Windows Azure Storage customers 100 percent for this disruption in their December bills.