Azure compute service disappears for 2.5 hours in West Europe region; cloud experts downplay possible Olympics connection.
Microsoft Office 2013: 10 Best Features
(click image for larger view and for slideshow)
The Windows Azure cloud compute service was unavailable to many users in Europe for 2.5 hours on Thursday, starting at 11:09 a.m. Greenwich Mean Time. The cloud compute service in the West Europe region is provided by Azure servers in Microsoft's Dublin, Ireland, data center and in an Amsterdam facility.
Microsoft spokesmen were reluctant to comment on why the service was unavailable shortly before lunchtime in Western Europe, even though there appeared to be no power supply interruption.
The loss of service access prompted a burst of messages on Twitter. "SoundGecko is currently unavailable due to data centre outage on Windows Azure. Apologies for inconveniences."
Azure storage services continued to be available and running workloads stayed up, regardless of whether users could access them or not, making the cause of loss of availability a different case from the Feb. 29 outage where a security certificate entered Azure's infrastructure without the ability to recognize it was a unique leap year date.
A previous outage affecting Amazon's EC2 service in Dublin last year was attributed to an interruption of power caused by a lightning strike. Amazon later backed off the lightning strike claim and said other power issues had cascaded into an interruption of service.
In Azure's case on Thursday, the constant availability of power and lack of a software culprit, such as the Feb. 29 one that downed several services, points to "more of an infrastructure issue" where some undetected single point of failure in a network switch or other device temporarily disrupted availability, said Chris Leigh-Currill, CTO of Ospero, a supplier of private cloud services in Europe.
Asked if interest in the Olympics, particularly the France vs. U.S. women's soccer match, could have caused spikes in network traffic, Leigh-Currill pooh-poohed the idea. The main surge in Olympic network traffic isn't expected until Thursday night and Friday. "Despite the British press's tendency to see Olympic traffic in everything, I can't imagine a large enough spike that would have affected the Dublin data center," he said.
Leigh-Currill said he watched the France vs. U.S. soccer match live at his desk off a live-stream feed and it continued uninterrupted without any problems.
Azure compute service was restored about two and half hours after it began at 1:33 p.m. Greenwich Mean Time.
"The recent Windows Azure service interruption in the Western Europe sub-region has been resolved... The duration of the service interruption was approximately 2.5 hours. Customers who have questions regarding this incident are encouraged to contact Customer Service and Support," said the restoration announcement.
The statement of resumption of service included Microsoft's regrets: "We apologize for any inconvenience this outage may have caused our customers." Microsoft said it is conducting diagnostics on the loss of service availability. "We will follow up as soon as we know more," said a spokesman. After the Feb. 29 outage, Microsoft issued a full post mortem explanation on the cause.
While Microsoft spokesmen said activities by users in Azure's Northern Europe sub-region continued uninterrupted.
Expertise, automation, and silo busting are all required, say early adopters of private clouds. Also in the new, all-digital Private Clouds: Vision Vs. Reality issue of InformationWeek: How to choose between OpenStack and CloudStack for your private cloud. (Free with registration.)
Multicloud Infrastructure & Application ManagementEnterprise cloud adoption has evolved to the point where hybrid public/private cloud designs and use of multiple providers is common. Who among us has mastered provisioning resources in different clouds; allocating the right resources to each application; assigning applications to the "best" cloud provider based on performance or reliability requirements.