From Sidekick to Gmail: A short history of cloud computing outages

This past week's Microsoft-T-Mobile-Sidekick data loss mess is the latest in a string of high profile cloud computing outages that have grabbed headlines over the past couple of years. Inevitably, the coverage of the initial outage (as well as cloud security breaches) is followed by explanations of why the outage happened (human error, network equipment, hackers, etc.) and analysis stories pointing out the pitfalls of putting your faith in the cloud.

Here's a short history of cloud computing SNAFUs:

Microsoft Danger outage: Contacts, calendar entries, photographs and other personal information of T-Mobile Sidekick users looks to be lost for good following a service disruption at Sidekick provider Danger, a Microsoft subsidiary. The amount of data and number of users affected wasn't disclosed by Microsoft or T-Mobile, but Sidekick support forums were buzzing with pleas from users looking for tips on how to restore their devices or get their data back.

Google Gmail fails…again:  When Google's Gmail faltered on Sept. 24, it wasn't down for more than a couple of hours, but it was the second outage during the month and the latest in a disturbing string of outages for Google's cloud-based offerings, including Google search, Google News and Google Apps over the past 18 months. Various explanations have been served up by the vendor, from routing errors to server maintenance issues. Some have come to Google's defense, saying that even though the company has had its share of outages, we are talking about mainly free services (you get what you pay for, in other words).

Twitter goes down…and yes, that's news:  While Twitter had been keeping its Fail Whale in hiding more often than not, a big Twitter outage that lasted throughout the morning and into early afternoon in early August had social networking types fuming. A denial-of-service attack was blamed for the problem.

eBay's PayPal crashes: The PayPal online payments system failed a couple of times in August, leaving millions of customers unable to complete transactions. A network hardware issue was fingered as the culprit for the outage, which lasted for between 1 and 4.5 hours, depending on how you look at it. It cost PayPal millions of dollars in lost business; it's unclear how much it cost merchants.

Rackspace pays up:  Rackspace was forced to pay out between $2.5 million and $3.5 million in service credits to customers in the wake of a power outage that hit its Dallas data center in late June. Rackspace, which offers a variety of hosting and cloud services for enterprise customers, suffered power generator failures on June 29 that caused customer servers to go down for part of the day. More disruptions followed and Rackspace kept customers up to date via its blog. 

Windows Azure test release goes down:  Early adopters of Microsoft's cloud-computing network Windows Azure suffered an overnight outage over a weekend in mid-March during which their applications being hosted on the network weren't available. This was only a test release of Azure, so observers noted that this obviously wasn't as big a deal as a production service outage. Separately, Microsoft also suffered a Hotmail messaging system outage in March. 

Salesforce.com kicks off the Year of the Cloud Outage:  As CIO.com's Thomas Wailgum reported in January, Salesforce.com suffered a service disruption for about an hour on Jan. 6 due to a core network device failing because of memory allocation errors.

Amazon S3 storage service knocked out:  We actually have to go back to summer of 2008 to find coverage of the last major Amazon S3 cloud network outage, which lasted for 7 to 8 hours and followed another outage earlier last year caused by too many authentication requests.

IDG News Service contributed to this story.