Twas the night before Christmas and all through the house, people were screaming and it wasn’t at a mouse. The in-laws were staring at the TV with care, in hopes that Netflix soon would be returned there. Alas, it wasn’t to be. Netflix wouldn’t be playing again until December 25th.
So what happened? It turns out it was still another Amazon Web Services outage. This one was centered at the Amazon Web Services (AWS) Elastic Load Balancers (ELB), in the US-East-Region1 data-center in Northern Virginia. Other cloud services went down as well, but let’s face it, on Christmas eve no one noticed much of the others. Curiously though, Amazon Instant Video kept going.
To be exact, trouble developed with the Elastic Load Balancers’ application programming interfaces (API)s. New load balancers would go up but they wouldn’t properly report their status to the overall ELB service. This, as Netflix users quickly found out, lead to “significant levels of traffic loss.”