Skype blames deficient networking algo for service disruption

Skype blames deficient networking algo for service disruption

Summary: Just received this statement from Skype about the current sign-on outage:Apologies for the delay, but we can now update you on the Skype sign-on issue. As we continue to work hard at resolving the problem, we wanted to dispel some of the concerns that you may have.


skype-logo4.jpg Just received this statement from Skype about the current sign-on outage:

Apologies for the delay, but we can now update you on the Skype sign-on issue. As we continue to work hard at resolving the problem, we wanted to dispel some of the concerns that you may have. The Skype system has not crashed or been victim of a cyber attack. We love our customers too much to let that happen. This problem occurred because of a deficiency in an algorithm within Skype networking software. This controls the interaction between the user’s own Skype client and the rest of the Skype network."

"Rest assured that everyone at Skype is working around the clock — from Tallinn to Luxembourg to San Jose — to resume normal service as quickly as possible."

Quick diagnosis, customer and media communication and of course, quick and enduring fixes are the benchmarks for how telecommunications companies manage service disruptions such as this,

Topics: Networking, Collaboration, Social Enterprise

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • I think is some windoze patch

    when M$ applies a patch that premium software stops working.
    A good reason to switch to Linux!
    Linux Geek
    • a typical troll response

      Company admits it was THEIR software that was the problem, and yet the trolls still come out of the woodwork and try to pin it on MS. Corporations don't bet the farm on trolls.
      Marty R. Milette
  • I don't get it -

    if [b]Skype[/b] ?loves [its] customers too much? to allow traffic to be disrupted by a cyber attack - over which they can have little direct control, why then do they allow this same traffic to be disrupted by a faulty algorithm in their networking software, for which they bear the sole responsibility ? I can accept that nothing is perfect in [i]le meilleur monde possible[/i], but this type of corporate spin does nothing to increase my confidence in [b]Skype[/b]'s management....


    PS : [b]Skype[/b] service does not yet seem to have been restored in my neck of the woods (Stockholm)
    • Reliability is based on learning from failure

      I think it's perfectly reasonable that they are able to protect themselves from
      intrusion better than they are from software error. Consider the fact that they are
      using established networking protocols which have been developed over decades.
      The security procedures and precautions for services on the Internet are well
      known. Skye software uses proprietary protocols - Skype is the only company
      using and developing their protocol, so the only time the software fails is when
      Skype itself fails.

      Contrast this to SIP, for example. This is a protocol suite developed by many
      people, with implementations by dozens of manufacturers and software projects.
      There are many more eyes looking at that code, and many more users finding new
      ways to break it.

      The benefit of using Skype is free "phone calls" to other people using Skype. The
      downfall is that it uses relatively immature technology. Expect failures like this as
      Skype grows further and further past its original design goals.
  • Some folks are complainers

    Digital networks go down. It is a fact of life. I work in the networking industry (manufacturers of networking equipment) and have seen even the most robust networks fail (yes, even top of the line network providers' networks). They get fixed and life goes on.

    The fact is, Skype is the best game in town for what they provide. If you don't like them, you can always change providers and hope for the best. But Skype is relatively inexpensive (and why it is the choice for many small businesses) and if you want more, you generally have to pay more. Personally I would rather pay much less for something that is 99% reliable than a lot more for one percent more reliability -- and even then there are no guarantees.
    • Out of touch - get real please

      Skype is either providing enterprise grade service or they aren't.

      If they are they should work with the 5 9's rule like the rest of the Telco's do. If they don't - then concede to being charlatans and get the heck out of the telephony market. Anybody who ever proposed use a sliding scale for reliability and tariffs never placed an emergency call.
      • false dichotomy

        You're basically saying if they can't be perfect they should quit completely. This is a fallacy of reasoning called the false dichotomy. Basically, you're saying there are only two choices: Provide essentially perfect service, or shut down completely.

        This is neither practical nor reasonable. Skype screwed up, and they will pay the price. Demanding their head on a platter is just piling on.
      • Shades of AT&T

        This reminds me of the time years ago when AT&T long distance service was interrupted for several hours due to a bug in their traffic routing algorithm. Skype obviously isn't the first, nor the largest telecommunications company to encounter a bug that they could not work out in their development systems.
      • re: out of touch

        What planet do you live on?

        99.999% of 24 X 365.24 is 5 minutes a year.

        I can't think of a single public service, or private one for that matter, that has had less than 5 minutes a year of down time. The includes my electric power, my land line, my cellphone, my cable tv or internet connection.
        Even at work where we run 30+ servers it takes MORE than five minutes to reboot one of them after they've crashed following patch Tuesday, or a phase change of the Moon.
        • Math wizard

          Regarding the attainability of this objective (which planet folks that refer to 5 9's live on) understand that 5 9's is an established legacy telecommunications standard - based on historical data from switched circuit providers - you could start with "99.999% availability has been used as a benchmark by the telecommunications industry in specifying high availability for many years." per "So you want 5 9?s?" - IEEE comment, also available at

          Regarding the issue with uptime with your pc or servers - my condolences - 5 9's comes at a price - perhaps one not necessarily contemplated or aimed at by those funding your IT budget? But then again you may not be handling 911 calls from those whose lives are in peril - for whom which the 5 minutes in question may be all that they have left?

          It's a price which has allowed some legacy providers to define the circuit dependability standards which Skype clearly missed here.

          Incidentally - your math is respectfully marginally out - that would be closer to 99.999% of 24 x 365.25 - a similar but not identical figure - to the extent that your point has to do with accuracy of figures consider it taken.
        • Agree

          Totally agree on that. Here at work in one of the world "top ten" carriers, this glamorous figure of <i>5 "9's"</i> is by far unrealistic.

          If we add to this the fact that VoIP transit on unreliable IP networks (Internet in this case) using active phones, based on servers with software (again the glamorous words <i>"fault tolerant", "mission critical" </i> systems at place) then Skype must forget assureing such high figures.

          But, have they ever assumed or guaranteed such reliabiity?
  • Skype

    Agree with those whom say isn't convincing. What's an algorhythm? They do not explain nor define.