Amazon issued a long technical explanation and apology for the outage that took Netflix down on Christmas Eve, without ever mentioning the rival online-video service by name.The outage lasted all night and part of Christmas Day.
Netflix was the most notable outage, but it wasn’t the only one. Amazon’s downtime affected other customers, like Salesforce.com’s Heroku. And Because Heroku is itself a cloud service that hosts other company’s apps, this mean that Amazon’s outage affected all kinds of companies.
It did not, however, affect Amazon’s own movie-streaming service that competes with Netflix, Amazon Instant Video.
On Saturday, Amazon explained that human error caused the initial problem. Someone inadvertently deleted some very important data. It promised this particular error would never happen again:
“We want to apologise. We know how critical our services are to our customers’ businesses, and we know this disruption came at an inopportune time for some of our customers. We will do everything we can to learn from this event and use it to drive further improvement in the ELB service.”
Because movie-watching is a big tradition on Christmas, it was Netflix’s outage that created the most fuss. Netflix fielded a long string of tweets on its customer service account, @Netflixhelps, where it noted that Amazon was to blame.
This, of course, prompted a lot of witty tweets in response like this one:
Dear Amazon Web Services: Please join us under this bus. Sincerely, @netflixhelps
— Clint Hoagland (@vsComputer) December 25, 2012
The service was mostly restored by 10:30 a.m. Pacific Time on December 25, but it left people wondering if Netflix was wise to trust Amazon’s cloud so thoroughly.
Last month, Netflix CEO Reed Hastings said that by the end of 2013, he wants Netflix to host 100% of its service on Amazon’s cloud, which would make Netflix the biggest customer for Amazon Web Services.