07/28/2007
TowleTech V.66
A series of power outages in San Francisco caused a massive web host to completely lose power for an hour Tuesday, taking down huge sites like Craigslist, Technorati, Typepad, Second Life, and many others. The outage, you may have noticed, also affected Towleroad. The power was restored in less than an hour, but it took all day for some sites to get back to fully-functioning. Since commercial-grade data centers promote themselves as having lots of back-up power solutions for extreme cases like this one, the failure was baffling. 365 Main, the faulty data center, says they are continuing to investigate what caused the total failure.
AT&T and Apple had an awkward situation early this week when both reported widely different numbers in the number of iPhones activated versus iPhones sold. AT&T claimed that a disappointing 156k iPhones were activated, sending Apple stocks dropping. But Apple recovered with reports that it sold 270k iPhones in the first 2 days. The reason for the discrepancy was unclear.
Netflix is struggling. Will Blockbuster choke them out of existence, or will the mail-order movie rental innovators make a comeback?
The $100 laptop, an initiative to bring affordable technology to children in developing countries, will finally begin production this year, with units becoming available for sale this fall. The organization behind the laptop, One Laptop Per Child, has fought against much criticism over the past 5 years, but has proposed an innovative design that is cheap, low power, and rugged.
Tech Video of the Week In light of the recent "crash" of many popular websites noted above, here is a video we showed you earlier in the week from the strangely prescient Onion News Network. Breaking News: All Online Data Lost After Internet Crash - Officials confirm that all online data has been lost after the Internet crashed and was forced to restart.
TowleTech is written by TR correspondent Daniel Williford.
Posted 1:00 PM EST by Daniel in TowleTech | Permalink
Like it?
Subscribe to FREE Towleroad daily headlines with our RSS feed!
RECENT STORIES:
Comments
You can follow this conversation by subscribing to the comment feed for this post.
Colocation of servers anywhere in San Francisco makes absolutely no sense when there are hundreds of other choices. Given that San Francisco could be devasted at any moment by a major earthquake, fiber optic lines, electricity and access to colo rooms in high rises could be unavailabe for weeks, months or longer. Unless you have a current offsite back-up or mirror site running in another city, your data and web business are at constant risk of vanishing forever. This is not the first time San Francisco based websites have gone down due to back-up power failure. This should be a wake up call to move to a safer more reliable location.
Posted by: Johnny Lane | Jul 28, 2007 1:24:24 PM
As usual APPLE was not prepared via their servers to have that many people utilize itunes to activate the Iphones. The fix didn't come around, via an updrade, until the Wednesday afterwards and either people go sick of 'em, or; and this is probably more accurate- people bought them to bust them open and reprogram them so they won't be so put out in the next upgrade that will leave 210K people wondering what they stood on line for. We remember what happened when we bought Nano too soon.
Posted by: lexxicuss | Jul 28, 2007 1:50:33 PM
Johnny - While having off-site backups is a good idea, this shouldn't really be a call for anybody to move their colo facilities from the SF bay area. The reason that they're there in the first place is because the bay area is one of the largest network hubs in all the world. It's where the bandwidth is. Of course a natural disaster might take out the hub, but if that is the case then it's very likely that network outages will pale in comparison to the other crises hitting the city. The issue here is that 365 Main should have been better prepared, and the affected sites should have had contingency plans.
Posted by: Vega | Jul 29, 2007 2:25:05 AM