News

VMware's Cloud Foundry Recovers from Outages

• By Jeffrey Schwartz
• 05/03/2011

VMware's fledgling Cloud Foundry service experienced two brief outages last week.

The outages pale in comparison to Amazon Web Services' four-day outage that concluded early last week. Amazon runs the largest cloud service which powers some large Web infrastructures, whereas Cloud Foundry is only a few weeks old and is still in beta. Also, the duration of the VMware outages was substantially shorter.

Cloud Foundry, introduced on April 12, is a cloud service that VMware is touting as the first open Platform as a Service, or PaaS. VMware offers its own hosted version of Cloud Foundry, aimed at letting developers build applications and run them on its service.

The first outage occurred last Monday, April 25, and lasted about 10 hours. Those most impacted were new customers who received their credentials the previous evening, said VMware's Dekel Tankel in a blog post explaining the two outages.

Last Monday's outage was the result of all eight cloud controllers losing connectivity to a storage subsystem. The root cause was a "partial outage of a power supply in a storage cabinet which impacted access to a single LUN," Tankel noted.

The next day's outage occurred when an engineer touched a keyboard, which caused a complete outage of the network infrastructure fronting Cloud Foundry, according to Tankel. It was Cloud Foundry's first compete outage. It lasted about 75 minutes.

"During this outage, all applications and system components continued to run," Tankel said. "However, with the front-end network down, we were the only ones that knew that the system was up. We take full responsibility for these issues and apologize to our users who were impacted by them. We can and will do better, having already learned from these incidents."