Sign in / up

back to article Amazon cloud still on fritz after 36 hours

Amazon's cloud is still on the fritz, a day and a half after the company first reported connection problems, latency issues, and increased error rates across the service. But on Friday morning, the company said that full service should be restored for a "majority" of users by the afternoon Pacific time. "We continue to see …

COMMENTS

House rules Send corrections
This topic is closed for new posts.
  1. Anonymous Coward
    FAIL

    Who was it?

    ...That said Cloud services were the best way to go to ensure data availability and continuity? Glad my mission critical servers are in-house....if a fire takes out our building, sure we go dark until we get new servers, but guess what? We don't have a building to do business in anyway! Network blip takes us off the web for hours? Good thing we have our servers in-house, eh?

    /rant

    6 2
  2. flying_walrus
    FAIL

    With 99.95 uptime...

    They're now promising zero downtime for the next 9 years

    5 2
    1. Geoff Campbell Silver badge
      Reply Icon
      FAIL

      You Fail, Sir.

      That's not what 99.95% uptime means. They've blown it for this quarter/year/whatever their measurement period is. Next measurement period, the clock is reset.

      GJC

      1 0
    2. Anonymous Coward
      Reply Icon
      Anonymous Coward

      With what guarantee?

      What are they willing to pay if they don't meet their SLA's. A lot of these public cloud companies don't provide any financial backing of their promises.

      0 0
      1. Geoff Campbell Silver badge
        Reply Icon
        Pirate

        @AC

        Absolutely - that's what the SLA should define. And an uptime with no service credits or refund for breaches is nothing more than a starting point for a really cool origami crane....

        GJC

        0 0
  3. Anonymous Coward
    Pint

    The need Power Systems....as in Power7

    Those x86 systems have MTBF of around 5 years and when you have 1,000 servers you are dealing with two failures every day.

    1 4
  4. Bill Neal
    Flame

    PSN?

    is the PlayStation Network on this cloud?

    1 0
    1. Anonymous Coward
      Reply Icon
      FAIL

      No - PSN down to anonymous DDoS attacks

      As title. Confirmed that The Xbox loving American hackers are responsible for the PSN DDoS attacks.

      Microsoft must be loving it, they seem to be using it as a PR stunt to promote XboxLive (conveniently forgetting their past 14 day Xmas outage).

      http://www.electronictheatre.co.uk/index.php/industry-news/7866-microsoft-react-to-playstation-network-outage

      “That being said, we are expecting Microsoft’s robust online network to see an increase of traffic from those gamers who own both systems. Being able to play their games via the Xbox LIVE network could make all the difference for some gamers, and the Xbox Nations event will allow all of them to do just that – whether they have a Gold subscription or not.”

      Seems Microsoft have forgotten about their 14 days of Xbox Live outage, and I don't recal Sony using it as a cheap PR stunt....

      http://www.engadget.com/2008/01/03/xbox-live-outage-day-13-still-up-and-down-still-preventing-fu/

      1 0
  5. Steve McPolin
    FAIL

    This is serious stuff.

    The entire next generation of needless programming projects is at risk if customers become skeptical of "the cloud". That would be a major financial setback for the industry.

    8 0
  6. amanfromMars 1 Silver badge

    Quantum Control System Leaping.. Function follows Phorm in Impulsive Drivers of Insatiable Desire

    So the information disappears ..... it is not the end of the world. And such petty losses really focus one's attention on what one needs old fading memory and failed historical pictorials for, whenever one is into building Future Post Modern CyberIntelAIgent Communities as Valid and Virtual, Viral and Venal, Venerable and Valiant InterNetworking Hubs for Alien Societies in Foreign Lands/Altered Time Zones/a Place of Strangers ...... although if the Future is not the Present or the Past, will only new information and intelligence be its Base Root Source.

    1 1
  7. WinHatter
    Pint

    The Java Paradigm knock off

    Corollary-1

    Ama-zone cloud : designed once fails everywhere.

    I know bugger all about their design but I'd assume, to cut costs, they copied-pasted their zones (infrastructure, hardware, hypervisors, OS & software). Ooooh shit-ngle point of failure.

    Corollary-2

    Hacked once hacked everywhere.

    May be some are happy to (help) see them fail. Ooooh Pooh again.

    0 2
    1. Ken Hagan Gold badge
      Reply Icon

      Re: designed once fails everywhere

      "I know bugger all about their design but I'd assume, to cut costs, they copied-pasted their zones (infrastructure, hardware, hypervisors, OS & software). Ooooh shit-ngle point of failure."

      Yup. Amazon are now blaming a "network event", hoping that no-one will notice that all their clouds responded to that event in the same way. Maybe if you and I keep very, very quiet about this, they'll get away with it.

      0 0
  8. Doug Glass
    Go

    Snakeoil

    All of it ... the whole bleeding cloud.

    5 1
  9. proto-robbie
    Alien

    Just a moment...

    ...The aliens aren't quite ready for us yet.

    0 0
  10. Anonymous IV
    Thumb Down

    "I wandered, lonely, as a cloud..."

    ... that promised utmost reliability.

    Wordsworth would be turning in his grave.

    1 0
  11. stefan 5
    Thumb Down

    The title is required, and must contain letters and/or digits.

    anybody whos lost data or lost websites e.t.c whatever.. serves you rite for being ignorant and using it in the first place. Cloud computing. what a bloody joke.

    1 0
  12. Anomalous Cowlard

    Reliable service is one thing

    ... but how about the ability of moving one's business elsewhere if that or other aspects are not satisfactory. I somehow very much doubt that the standards allowing this to happen have been devised/implemented (or ever will be) as this is a) very difficult and b) losing lock-in would mean actually having to compete.

    1 0
    1. Tom Chiverton 1
      Reply Icon

      dont be silly

      Don't be silly. There are many cloudy places that will accept a standard J2EE .war file.

      0 0
  13. Dave N
    WTF?

    To restore confidence...

    ...Amazon need to release technical details ASAP, and explain how they're going to prevent this happening again.

    0 0
  14. mmouse19

    behind the scenes at AWS

    http://www.youtube.com/watch?v=m3wrBFuGK2A

    1 0
  15. Jeff 11
    Flame

    The cloud...

    ...good for apps and data you don't need to care about. Less so for ongoing mission-critical business processes.

    Ultimately "moving to the cloud" is all about outsourcing your data and processing to some faceless entity that's trying to do everything as cheaply and generically as possible. That's generally a mistake you'll come to regret in business, IT or otherwise.

    2 0
This topic is closed for new posts.

Other stories you might like