Amazon EC2 Goes Dark In Morning Cloud Outage
Andrew R. Hickey
According to AWS's status dashboard and service updates, the company began investigating connectivity and latency issues around EC2 instances in its U.S. East Coast region at 4:41 a.m. Eastern. It appears the issues only affected Amazon's North Virginia data center.
A short time later Amazon confirmed the EC2 cloud issues.
"We can confirm connectivity errors impacting EC2 instances and increased latencies impacting EBS volumes in multiple availability zones in the US-EAST-1 region," Amazon wrote at 5:18 a.m. "Increased error rates are affecting EBS CreateVolume API calls. We continue to work towards resolution."
The issues persisted. In a 5:49 a.m. update, Amazon said it continued to see connectivity errors impacting EC2 instances. Additionally, Amazon was "also experiencing delayed launches for EBS backed EC2 instances in affected availability zones in the US-East-1 region."
As of 6:20 a.m., service began to recovery, and by 8:02 a.m. Eastern AWS said "latency has recovered for a portion of the impacted EBS volumes. We are continuing to work to resolve the remaining issues with EBS volume latency and error rates in a single Availability Zone."
Come 9:09 a.m. Eastern, API errors and volume latencies in the affected ability zones remained, Amazon said. And at just before 10 a.m., Amazon wrote "there has been a moderate increase in error rates for CreateVolume. This may impact the launch of new EBS-backed EC3 instances in multiple availability zones in the US-East-1 region. Launches of instances store AMIs are currently unaffected."
Amazon did not respond to a request for comment as of press time.
The outage affected a host of big name Amazon EC2 cloud customers, including popular Web sites like Foursquare, HootSuite, Quora and Reddit.
Early Thursday morning, Reddit visitors were met with the message: "Amazon is currently experiencing a degradation. They are working on it. We are still waiting for them to get to our volumes. Sorry."
Meanwhile, Quora, a newcomer social site that lets people ask and answer questions said "We'll be back shortly, we hope. Sorry, it sucks for us too. We'd point fingers, but we wouldn't be where we are today without EC2." Quora's Web site was offline as of 10 a.m.
Thursday wasn't the first time Amazon's cloud computing services suffered outages. As more users have moved to the cloud, cloud outages become more common.
Cloud outages have also raised questions around cloud support, with vendors and solution providers working to determine who is the first to take the call when the cloud experiences a hiccup.
Earlier this year, Amazon Web Services adjusted its cloud support options and lowered its support pricing structure.