RESTON Va., December 25, 2012 – After an abrupt and severe East Coast outage that commenced around 4 p.m. EST on Christmas Eve, December 24, it appears that Netflix’ streaming service is coming back on line. As of approximately 1:30 p.m. EST today, help functions as well as streaming movies and TV shows were once again available to those users, like this writer, who were affected by the outage. It was reportedly caused by a malfunction in elements of the cloud architecture implemented by Amazon servers that host Netflix’ services in the vicinity of the crucial MAE East IXP (Internet Exchange Point) clustered primarily in northern Virginia, in and around McLean, Reston, and Vienna.
A substantial number of DC area and East Coast customers were without the service since yesterday afternoon, with the outage affecting “many but not all devices” according to a statement by Netflix, which tweeted around 11:45 a.m. EST that their service was “back to normal streaming levels.”
When the streaming service went down yesterday, the company was quick to apologize via Twitter and other available outlets. “We’re sorry for the Christmas Eve Outage. Terrible timing! Engineers are working on it now,” the company tried to explain as thousands of disgruntled holiday Netflix users flooded the Twitter-verse with negative comments.
Various engineering sources as well as Netflix itself blame Amazon Web Services servers for the problem which has been pinned on portions of the company’s cloud architecture that apparently aren’t as flexible and robust as they need to be to handle sudden, massive bursts of traffic that services like Netflix are likely to encounter during major events or holidays like this one.
The outage was somewhat peculiar in that it appeared to affect some customer devices but not others. Shortly after the problem began to spread, Amazon started reporting that some of its “Elastic Load Balancers” (ELBs) were down, elements in its cloud architecture that are allegedly supposed to handle the kind of sudden traffic surges that apparently overwhelmed them on Monday.
According to CNET, Netflix spokesman Joris Evers “noted that ‘streaming was available again for the majority of our members late on Christmas Eve Pacific Time.’” That was not the case for many DC area customers, however, who were unable to connect to the service until just after noon today.
While the current outage is a black eye for Netflix, coming as it does on a major holiday, the problem may be more of an issue for Amazon, which serves a number of major sites besides Netflix, many of which also experienced serious downtime.
Outages lasting an hour or two as well as sudden interruptions of streaming service have become more common for Amazon’s commercial network customers like Netflix over the last few months, indicative of a growing problem in the handling of peak service that hasn’t been adequately addressed by the company’s engineering team.
This issue also points to the continuing fragility of the Internet itself, which is increasingly vulnerable to international hackers.
UPDATE: According to a Reuters report, the Christmas Eve-Christmas Day Netflic outage not only had a serious impact on subscribers in and around the U.S. East Coast and environs. “The outage impacted Netflix subscribers across Canada, Latin America, and the U.S., and affected various devices that enable users to stream movies and television shows from home, Netflix spokesman Joris Evers said,” according to the report. “Such devices range from gaming consoles like the Nintendo Wii and PlayStation 3 to Blu-ray DVD players.”
Reuters observed that “more than 27 million” Netflix subscribers “are in the Americas region that was exposed to the outage and could have potentially been affected.”
Netflix confirmed that the massive outage was the result “of an outage at an Amazon web services’ cloud computing center in Virginia. It apparently started around 5:30 p.m. EST on Monday, Christmas Even, and was fully restored at around 11:00 a.m. EST Christmas morning.
According to Reuters, “The event marks the latest in a series of outages from Amazon web services, with one occurring in April of last year that knocked out such sites as Reddit andFoursquare.”
Read more of Terry’s news and reviews at Curtain Up! in the Entertain Us neighborhood of the Washington Times Communities. For Terry’s investing and political insights, visit his Communities columns, The Prudent Man and Morning Market Maven, in Business.
Follow Terry on Twitter @terryp17
This article is the copyrighted property of the writer and Communities @ WashingtonTimes.com. Written permission must be obtained before reprint in online or print media. REPRINTING TWTC CONTENT WITHOUT PERMISSION AND/OR PAYMENT IS THEFT AND PUNISHABLE BY LAW.