August 26, 2013
Amazon Outage Takes Down Instagram, Vine And Others
Michael Harper for redOrbit.com - Your Universe Online
Amazon’s Web Services (AWS) experienced an outage yesterday, taking down some of the web’s most popular sites. Airbnb, Instagram and Vine users all had difficulties accessing these sites yesterday as AWS began having issues with their Elastic Compute Cloud (EC2) offerings, relational database and load-balancing services. Amazon offers their EC2 services to host web applications.
“We know many of you are having trouble loading Instagram. We identified the issue and are working to fix it ASAP,” wrote Instagram on its Twitter feed. Vine followed with: “We're aware of some issues affecting our servers and are working to address them now. Thanks for your patience and hang tight!"
The AWS outage may have been relatively short lived, but it’s absence was notable. Many websites large and small use AWS to host their applications and web services. As such, AWS is the backbone for some of the world’s most popular sites. While the issue was ongoing, Amazon kept users up to date with their process for finding the problem and resolving it. The company said they found that one of their networking devices was failing, but they weren't quite sure why.
"The networking device was removed from service, and we are performing a forensic investigation to understand how it failed," said AWS in a statement. “We are continuing to work on a small number of instances and volumes that require additional maintenance before they return to normal performance."
This failure caused a degradation in some of their Elastic Block Store (EBS) storage used in partnership with EC2. This resulted in API requests returning with error messages. The failure took place in their US-EAST data center, and while the company normally keeps track of the status of their cloud offerings online, the AWS Health Dashboard reported all systems were operational during the outage. At the time of this writing the Dashboard lists that all things were operational on August 25, but not without a brief caveat.
At 5:21 PM EST, the Health Dashboard read: “We have identified and fixed the root cause of the performance issue. EBS backed instance launches are now operating normally. Most previously imp aced volumes are now operating normally and we will continue to work on instances and volumes that are still experiencing degraded performance.” AWS says normal service was restored before 9 PM EST.
Instagram, Netflix and others, like Foursquare and Pinterest, have seen their services grind to a halt before as a result of an AWS outage. Perhaps the most memorable disruption was the Netflix Christmas Eve outage of 2012. As families across America gathered in front of their televisions to celebrate the holidays, they were unable to access Netflix on their computers, tablets or Xbox 360 units. Amazon rang in the new year explaining why Netflix had gone down on December 24th, saying one of their developers accidentally deleted a portion of data used to keep their services up and available.