It’s hard to believe July is finished and we’re on the back end of 2012. Taking a look at the first half of the year, here is our take on four of the most impactful server outages:
#4 Missing Microsoft
Tech giant Microsoft is usually hard to miss. Its Windows Azure cloud platform, however, has been anything but easy to locate after experiencing two separate outages this year. The first occurred on February 29 when the service was down for nearly a day. The company first attributed it to an April Fool’s outage; users were not amused. Microsoft credited customers’ accounts 33 percent. In tandem with the most recent Twitter outage that lead up to the Olympics on July 26 (see #2 below), Azure users experienced yet another outage. Microsoft apologized to for the outage and said the incident was caused by a “misconfigured network device.”
#3 Auction Outage
For online retailers, maintaining uptime on their website is vital to a thriving business. Unfortunately for eBay, one of the largest and most well-known auction and shopping sites, their site was down for several hours on April 26. eBay did offer some credit to vendors and customers, but who’s to know how many potential buyers or sellers didn’t wait around and sold or purchased items elsewhere?
#2 Double Down for Twitter
Using social media networks London Olympics athletes and fans communicate with one another, sharing Olympiad moments like never before. However, during the week leading up to the Olympics, Twitter was down for at least an hour on Thursday, July 26. Despite speculation, the outage had nothing to do with the IT load of the Olympics. According to official statements from Twitter, the outage occurred following the failure of two – count them, two – data centers. Twitter explained after the first data center went down, the site was to be picked up by another data center, however, instead of picking up the load, the backup center went down as well. Twitter apologized for the outage, and users have returned to tweeting without interruption.
#1 Low Clouds Brings Amazon Storm
Amazon seems to be at the eye of the cloud hurricane, except their world is anything but calm. After falling victim to two outages lasting a total of eight days in 2011, the online retailer had a lot of ground to make up with customers. As one of the most devastating storms of 2012 slam the mid-Atlantic region in June, the effects of the storm rippled through much of the U.S. and world as Amazon’s Northern Virginia data center also took a hit. Two weeks later, that same data center went down again.
What do you think of these outages and the lessons each company took away? Do you think we’ll see these same organizations fall victim to downtime again or do you think they will have learned their lesson and move forward with implementing proper high availability systems?