Another Amazon Web Services Outage Takes Out Top Websites

Reddit, foursquare and Pinterest ware among the large sites hit yesterday thanks to escalating issues at Amazon Web Service’s data centre in northern Virginia.

It marks the third time major outages have occurred as a result of Amazon problems in that region in the last five months. The biggest cloud provider in the world saw storms hit its northern Virginia data centre in July this year, disrupting power and taking out major services, including Netflix.

In June, the same data centre lost power, thanks to multiple backup failures.

Three-fold Amazon Web Services fail

Three operations were experiencing technical difficulties yesterday and into today. The biggest fire Amazon had to put out was in its Elastic Block Store (EBS) component, which forms part of its hugely popular Elastic Compute Cloud (EC2).

Around 6.30pm in the UK, or 10.38am PDT, Amazon, on its service health dashboard, reported “degraded performance for a small number of EBS volumes in a single Availability Zone in the US-EAST-1 Region”. At 4:42pm PDT, its fixes kicked in, as customers were able to launch new EC2 instances in the affected zone.

Its last post on the issue was at 10.54pm when AWS said it had “completed recovery of nearly all affected load balancers”.

“We will continue to work to restore IO for the remainder of volumes and will reach out via email to affected customers that own those volumes should action be required on their part. Volumes affected earlier in the day are continuing to re-mirror (which we expect will take several more hours) and while this process continues, customers may notice increased volume IO latency,” it added.

AWS’ Relational Database Service (RDS) was also disrupted in northern Virginia, with “connectivity issues and degraded performance for a small number of RDS DB Instances”.

The last update was at 2.25am PDT, when Amazon said: “Our recovery process to bring remaining RDS instances back on-line in the affected AZ is continuing at a steady pace. Customers can launch new database instances.”

The third service to be hit was the Elastic Beanstalk Platform-as-a-Service, which was “experiencing elevated API failures and delays launching, updating and deleting Elastic Beanstalk environments”.

Amazon has not responded to a request for comment on the problems, as it continues to investigate the root causes.

A host of major websites were reportedly affected, including Reddit, Foursquare and Pinterest.

Can you rely on your cloud knowledge? Take our quiz!