Netflix has restored its streaming video services, after Christmas outages related to problems with Amazon's online computing platform.
Netflix users were unable to stream movies and TV shows from some devices on Monday, with many posting complaints on the company's Twitter feed and online forums like Reddit. The company repeatedly told users via Twitter that it was working with engineers from Amazon Web Services to fix the issues, then said Tuesday it had resolved most of them.
"Our team worked with AWS through the night and issue was mostly resolved late Christmas Eve!" Netflix posted on its Twitter feed.
A public status page for Amazon's online platform said all services were operating normally as of Tuesday, but showed records of earlier problems along with apologies to customers for technical issues. The problems appeared to be with Amazon's load balancing and application deployment services from its servers based in North Virginia.
The outages at Amazon affected other downstream customers as well. Salesforce.com's Heroku, which provides users with an online platform of its own, posted a status message saying it had technical issues related to AWS, as did Scope, an online social hub.
AWS is Amazon's cloud platform, which charges websites and service providers for using its software and data centers to run their databases and online applications. The service, developed based on the infrastructure Amazon uses for its own online retailers, is largely invisible to end users when it runs normally.
The outage at AWS is the latest in a recent series of technical mishaps on the platform. Last month, a bug related to new hardware caused an outage that lasted over 12 hours, bringing down popular sites such as Reddit, Imgur and Heroku. In June, summer storms caused power outages that affected services at Netflix, Instagram and Pinterest.
Amazon has voluntary refunded its customers for downtime in the past, even when not required to do so contractually. The company has been criticized for its service level agreement (SLA) for AWS, which imposes strict guidelines in how customers must build their online systems, mandating terms such as having them run across regions supported by separate data centers to avoid outages.