SoylentNews is people
SoylentNews
Skype Outage Hassles Millions of Users
Multiple sources are reporting that Microsoft's popular messaging and voice chat application Skype has been hit by a severe service outage.
The problem seems to be a bug in the application which affects users from seeing each other. Microsoft has stated that they have found the cause of the problem and is slowly working to restore service, however at the time of this writing it seems that a number of global users are still affected.
Link to official Skype status: Skype Heartbeat.
I've been planning to move away from Skype for a while, but its popularity is making it somewhat difficult (akin to the Facebook dilemma: everyone uses it, why switch?) and this outage has underscored the need for redundancy, especially for some things that users view as essential for connectivity. What other similar applications can Soylent recommend?
AWS knocks Amazon, Netflix, Tinder and IMDb offline in MEGA data collapse
Amazon's Web Services (AWS) have been hit by a monster outage affecting the company's cloudy systems, bringing many sites down with it in the process.
The service disruption has hit AWS customers including Netflix, Tinder and IMDb, as well as Amazon's Instant Video and Books websites.
The outage may also explain Airbnb's current service woes. Airbnb is an AWS customer.
(Score: 4, Insightful) by mendax on Tuesday September 22 2015, @06:22AM
These kinds of outages could doom cloud computing... perhaps. However, as everyone here who knows anything about the history of computing, system failures have been occurring since the days of the ENIAC. Computers and their components break. The issue with cloud computing is that when something fails, the failure can potentially be catastrophic if not fixed quickly. However, cloud computing, virtual machines, auto-replication of data, and hardware redundancy can make this downtime minimal. I suspect that when the history of cloud computing is written, it will turn out to be far more reliable than colocation or hosting one's own servers.
Of course, cloud computing makes computing more of a commodity that can be consumed. As many of you know, this site is hosted in the cloud somewhere. It makes it expansion as it is needed a much easier proposition, and, of course, no servers needed to be found and put in a colo facility somewhere.
It's really quite a simple choice: Life, Death, or Los Angeles.
(Score: 0) by Anonymous Coward on Tuesday September 22 2015, @06:46AM
> As many of you know, this site is hosted in the cloud somewhere
Really? I thought soylent ran on dedicated servers. I guess that still "the cloud" in the most general sense.
But when I hear "cloud" I think VM that can be software migrated and dynamically resized/provisioned.
(Score: 2) by kurenai.tsubasa on Wednesday September 23 2015, @12:01AM
Provided a Dell hardware RAID hasn't just had a brain fart after a UPS failure, went to “repair” the RAID with the wrong parameters, and promptly fucked your VM up. True story.
That being said, I'm happy to report complete satisfaction with Linode, even though I only use my server in the clouds for hobby projects, so I don't have experience with a site even at Soylent's level. They even allow you to provision at multiple different physical data centers to control for the above scenario, and they're very prompt about reporting outages. In, what has it been, 7 years or so, Linode has never lost a copy of my VM.
(Score: 3, Interesting) by frojack on Tuesday September 22 2015, @06:48AM
What percentage of these outages (which are never adequately explained) do you suppose are actually hacking attacks?
American Airlines was down last week as well. That's like the third airline outage this summer.
Nobody wants to admit a hacking, but "technical difficulties" are apparently ok.
No, you are mistaken. I've always had this sig.
(Score: 0) by Anonymous Coward on Tuesday September 22 2015, @06:59AM
Man, you find conspiracies under every rock.
The internet is freaking enormous nowadays. A couple of high-visibility outages a month is really low volume.
Besides, hacks are publicly reported all the time, they just aren't sexy enough to make the news unless its credit card numbers or cheating husbands. If nothing else, the think-tanks that are pushing for cyberwar make a point of collecting and summarizing that stuff on a regular basis.
(Score: 2) by mendax on Tuesday September 22 2015, @08:21AM
Oh, it wouldn't surprise me if many if not most of them are hacking attacks. Cloud computing infrastructure properly implemented should be pretty resilient, with few single points of failure, unless there is a hacker attack.
It's really quite a simple choice: Life, Death, or Los Angeles.
(Score: 1) by SanityCheck on Tuesday September 22 2015, @11:11AM
Please, if these were hacking attacks the companies would be falling over themselves to proclaim "SEE IT WASN'T OUR FAULT." More likely it's someone's incompetence. Just like when United Airlines was down for an entire day, I had the inside scoop that someone installed a mis-configured router that was routing traffic in an infinite loop between Chicago and somewhere in Louisiana. Considering what United pays their IT "contractors" (i.e. slaves, which make $20 an hour in some cases with no benefits working through a sub contractor), I am more surprised their systems work at all!
(Score: 2) by frojack on Tuesday September 22 2015, @06:39PM
Please, if these were hacking attacks the companies would be falling over themselves to proclaim "SEE IT WASN'T OUR FAULT."
I've actually never heard any company make that claim, because everybody knows its STILL their fault.
But I've seen many companies be very circumspect about outages, and the reasons for them, and then fess up months later that they were hacked.
This is the third ground stop American has had this year alone.
No, you are mistaken. I've always had this sig.
(Score: 3, Insightful) by bradley13 on Tuesday September 22 2015, @09:24AM
Are their occasional outages? Yes, certainly. I don't see why these are any more catastrophic than any other infrastructure failure. If my company's servers are down, it doesn't really matter whether they were in the cloud or in the back room closet.
The earlier poster suggests running your own infrastructure, but he neglects to pose the key question: will you have fewer outages? Are you sure?
Even you would have higher reliability, you then have all the additional time, effort and cost of actually running your own infrastructure. Those costs are considerable, most especially the salaries of the people doing it. Is the extra reliability (if there is any) worth that cost?
Of course, there are other arguments to be considered as well, for example, security. I would never put anything on a cloud service in the USA, for example, but for the moment I do have things on AWS in the EU.
Everyone is somebody else's weirdo.