SEARCH

Entries in Outages (2)

Tuesday
Oct302012

Hurricane Sandy Takes Down Gawker, Huffington Post and Other Websites

Photos.com/Thinkstock(NEW YORK) -- It was inevitable. With power out along the Eastern Seaboard for millions of people, the data centers and servers powering many websites were also hit, knocking popular websites offline for hours.

The biggest casualty? Datagram, the Internet service provider based in New York City that powers news sites like Gawker.com, Huffingtonpost.com and Buzzfeed.com. When its servers went down on Monday night due to flooding, the sites it powered went down with it.

"Unfortunately, within a couple hours of the storm hitting Manhattan's shores, the building's entire basement, which houses the building's fuel tank pumps and sump pumps, was completely filled with water and a few feet into the lobby," Datagram said in a statement on its site. "Due to electrical systems being underwater the building was forced to shut down to avoid fire and permanent damage."

As a result Gawker, owner of sites such as Gawker.com, Gizmodo, and Jezebel, also went down late Monday night. "We're continuing to work on our servers and will be back online as soon as is possible. We miss you already. Stay dry," Gawker tweeted from its account last night.

Gawker switched to Wordpress and Tumblr, different website and blogging platforms, as backups for its site. Gizmodo has been live blogging the aftermath of Sandy on its Sandy 2012 Emergency Site.

"While we're obviously disappointed with Datagram, our priority has been getting back online for our readers with an alternate publishing platform, which we've now done with all sites, thanks to Tumblr," Scott Kidder, Gawker's Executive Director, Operations, told ABC News.

Buzzfeed.com and The Huffington Post were also affected by Datagram's outage. Buzzfeed, a site that has surged in popularity in the last couple of months, was able to recover its full site, though, more quickly than others were.

"Elements of BuzzFeed's site and many story pages are back online, thanks to a Content Delivery Network, Akamai, which hosts the content at servers distributed around the world," Buzzfeed wrote on its blog Monday.

"Two key things helped BuzzFeed recover: After Hurricane Irene last year, BuzzFeed commissioned an offsite datacenter that replicates everything in near real-time. More recently, the site started using Akamai to cache content. That means that when Datagram was offline, the site and its pages should have stayed up — and many did," Buzzfeed's Matt Buchanan said in a post on Buzzfeed's FWD tech site.

Similarly, The Huffington Post had a backup server in Newark to rely on.

"Between Monday night and Tuesday morning, HuffPost was accessible via a temporary site -- status.huffingtonpost.com -- and writers and editors relied on Tumblr, Facebook and Twitter to post stories and information during the storm," the Huffington Post said on its site following the outage.

MarketWatch also went down for a brief period of time, but it is unclear if that was a result of Datagram's outage.

On Tuesday afternoon Datagram had reported that Consolidated Edison and city workers were helping to restore the services and that there were "at least five pumps pumping water from the basement into the street."

While Datagram will eventually come back online, Kidder added that Gawker will be speeding up its plans to have a second data center. "We -- as other publishers -- had counted on Datagram's ability to withstand anticipated natural disasters, which seems to have been misplaced."

Copyright 2012 ABC News Radio

Friday
Oct262012

A Tough Week for the Internet as Major Sites Experience Outages

Photos.com/Thinkstock(NEW YORK) -- Last week it was YouTube for a few minutes, but this week it was more than just the streaming video service that wasn't available to Internet users.

Earlier this week, Amazon's Web Services went down causing sites that rely on Amazon's servers to go down too. Popular sites like Reddit, Pinterest, and Foursquare all experienced outages as a result. The outage lasted for a few hours on Tuesday, and naturally many took to Twitter to complain about the fact that people couldn't get to their services. Twitter itself experienced its fair share of outages.

On Friday a separate outage occurred. Google's App Engine, which powers other sites, along with Dropbox and Tumblr experienced outages. The outage lasted close to two hours for many of the services.

"At approximately 7:30 am Pacific time this morning, Google began experiencing slow performance and dropped connections from one of the components of App Engine. The symptoms that service users would experience include slow response and an inability to connect to services," Google wrote on its site.

Similarly Tumblr tweeted about the outage: "Tumblr is experiencing network problems following an issue with one of our uplink providers. We will return to full service shortly." Two hours later, Tumblr tweeted that the errors had been fixed and it was back online.

"It used to be back in the day, four or five years ago, systems weren't dependent on each other. But now even standard websites -- the things people go to all the time -- are made up of 50 or 100 services that are serving ads and tracking information," Brian Gracely, a Cloud computing expert and editor of CloudCast.net, explained to ABC News. "If one of the big services or an Amazon or Google goes down it can affect hundreds of other services."

Tuesday and Friday's outages don't appear to be related, but according to the Internet Traffic Report, traffic across the web in North America declined Friday. The Next Web points out that the same report shows that there was a loss in packet data, which measures reliability of Internet connections.

These issues do not appear to be weather related either. However, many Internet providers will be preparing as Hurricane Sandy makes its way to the East Coast.

"These companies, like Google and Amazon, run the equivalent of what used to be 20th century factories. They are really large and occasionally they have an outage because they have a power failure or weather issues," Gracely said. "It happens periodically, and it used to happen more than we knew, but nowadays we are so connected we know about it more," told ABC News.

Copyright 2012 ABC News Radio







ABC News Radio