10 biggest failures in the history of the Internet
Lightning, cartoons of the Prophet, the death of the King of Pop, difficulty in Amsterdam and other major causes of accidents on the network.
Last Friday, the home page and all services "Yandex" for a few hours stopped working. Internet outages for many a long time stood in a row with power outages or water: without it, it is very difficult to live.
Small and short-term disruptions in the work of most Internet companies are frequent. Forbes magazine has decided to recall the truly major Internet blackouts.
When: August 16, 1996
How long: 19 hours
More than 6 million customers the most popular ISP in the United States almost a day lost access to the Internet and AOL services due to the fact that its data centers were carried out repair work. CEO and founder of AOL Steve Case has publicly apologized and opened on the day of free Internet access to the company's clients.
AOL, which was originally called America Online, after the incident began to regularly call America Offline, and he entered the day of the failure of the Internet history as "Black Wednesday". In addition, AOL blackout provoked an avalanche of articles in the press about people's dependence on "World Wide Web".
When: July 15, 1998
How long: one day
Ebay failure in June 1998 became the first in a series of protracted site falls. Then Ebay has suffered a loss of $ 3 million and was forced to resume all broken auctions. During the year, the online auction service has experienced about 10 such cases. Every time the site blamed management failures in the software, the outdated architecture of the data center.
When: August 16, 2007
How long: two days
The first major blackout, which lasted about two days, a Skype experienced in August 2007: users could neither call or send messages. The reason - a problem in the algorithm to work with the network software. At that time, the market capitalization of Skype owner, service eBay online auctions, for the first day of failure fell by $ 1 billion.
In December 2010, Skype ceases to work again - this time for about three days. As a result, callers whose calls were interrupted at the moment of collapse, were presented with the opportunity to make a free call, and the owners of premium accounts services have been extended by one additional week.
Six months later - in late May and early June, and then - again, Skype users denied access to the service.
When: February 24, 2008
How long: two o'clock
In winter 2008, the authorities of Pakistan have decided to block their populations access to YouTube. The reason for the lock become featured on video sharing videos, "insulting Islam", in particular, the famous Danish cartoons of the Prophet Mohammed. However, some observers thought that really scared Islamabad that YouTube used as a repeater opposition materials.
Fifty Pakistani providers obediently complied with the requirement of the authorities, but in the end turned out to be disconnected from the video sharing almost the entire globe. Due to a series of errors, most providers of the planet began to redirect all traffic destined for YouTube in Pakistan, that is the only place where it was banned.
MobileMe (Apple service)
When: July 2008
How long: a few weeks
MobileMe was supposed to link together all of the devices Apple family, allowing the wearer to conveniently synchronize their contacts, important documents and photos. However, immediately after the launch of the service began to act up. In order to fully remedy the situation, Apple took about a month, but MobileMe's debut was totally ruined.
Apple MobileMe gave each user a free subscription for a month, Steve Jobs broke an angry letter, in which he called MobileMe "not meet the high standards of the company," and sacked the manager responsible for the project. Later, Apple admitted that she launched a crude product, and its servers were not prepared for the influx of traffic.
In June 2011, the head of the Jobs introduced a new service iCloud, MobileMe failure naming and announcing its closure.
When: August 11, 2008
How long: two o'clock
To access Gmail in August 2008 lost almost all of its 100 million users, including corporate customers and employees of Google. Cboy was caused by malfunctions in the contacts system.
Dissatisfied users began to complain about Gmail in your Twitter feed. According to some estimates, posts on Twitter about the failure mail client appeared at a rate of 100 pieces per minute.
Two hours later, the service was restored.
When: June 25, 2009
How long: One hour
One of the most famous and demonstration failures in Twitter work occurred on the day of Michael Jackson's death. An explosion of interest in the death of the King of Pop has led to a drop in service: a few minutes Twitter users sent tens of thousands of one way or another connected with Jackson messages. On Twitter restoration took about an hour and a few hours spent on the resumption of smooth operation. Twitter fails so often that it appears in these cases whale image you unsuccessfully try to save eight orange birds (fail whale), the fame of its author Ying Liu.
When: September 23, 2010
How long: three o'clock
A week after Forbes put the founder of Facebook Mark Zuckerberg over Steve Jobs' list of the richest Americans, social networking has experienced "the worst crash in its history." By the fall of 2010, the number of registered Facebook users exceeded 500 million, and the fall of the social network was a shock, and the main news of the day. Five of the top ten trends on Twitter have been devoted to this topic.
Facebook management politely apologized and explained in detail the reasons for the incident: an automatic system, which aims to identify incorrect settings, the changes considered a mistake made in the basic network database, so began a massive correction "errors", and the server were overloaded with hundreds of thousands of requests per second.
When: April 21, 2011
How long: up to four days
Failure of the "cloud" service Amazon Elastic Compute Cloud (EC2), yielding to rent computing power of the company, led to the temporary collapse of a number of popular sites and services, for example, Netflix, Zynga, Foursquare, Quora and Reddit. The reason was the wrong network settings, triggering overload marshuritizatorov. Customers began to get access to EC2 12 hours after the incident, but to eliminate all the effects took about four days. In early August, a lightning strike hit the data center Amazon, located in Dublin. A fire broke out and EC2 has failed again. If the first time is mainly affected American Amazon customers, now fails touched Europe.
When: August 19, 2011
How long: three o'clock
One of the biggest failures in the history of "Yandex" has occurred due to an error on the router, which is located in the new data center of the company in Amsterdam. As a result, a three-hour stop working not just search, but all the other services - mail, maps, weather, web analytics, and so on..
While representatives of the company explained the incident "technical difficulties" on the internet crept rumors of DDoS-attacks and fire in the data center. But these assumptions have been refuted.