21 Aug 2007
VoIP provider Skype has claimed that the outage that crippled its service last week was down to a surge of users logging on after installing Microsoft's latest security update.
The crash caused a two-day outage of the service and prompted speculation that the VoIP provider had fallen victim to a denial-of-service attack.
Further reading
Skype spokesman Villu Arak said in a posting to a company blog that, while the problem was caused by a system crash, it was not due to a deliberate malicious effort.
Arak explained that the outage arose when a large number of users installed Microsoft's August security update.
After restarting their computers the users all attempted to log in to the Skype service, causing an overload on the components of the service that handle peer-to-peer connections.
A "self-healing" feature would normally intervene and allow the system to correct or compensate for the increased load. But the feature failed to activate and the result was a chain reaction that took down the entire service.
"The issue has now been identified explicitly within Skype," Arak declared. "We can confirm categorically that no malicious activities were attributed and that our users' security was not, at any point, at risk."
Arak said that Skype has already patched the code responsible for the crash.
He did not say that the system would be able to weather a similar flood of traffic in the future, but did pledge that users would not be "similarly effected" the next time the issue arises.
Latest stories from Software
Related articles
Related jobs
Poll
What is the most important IT priority for your company this year?
Connect with V3.co.uk
This paper focuses on a series of best practices and techniques for development teams looking to improve their software development processes
Why good data management at all levels is essential in the modern business (video, 6mins)
Field/Site Engineering Manager/Leader Brief: Polar...
Product Manager, Open Repository (ref:BMC/PMR) End...
Java/J2EE Software Developer/Programmer - Dotcom/ eCommerce...
Field/Site Engineering Manager/Leader Brief: Polar...
Keep up to date with the latest products, services and technologies from the world's leading IT companies. IThound.com brings you over 2,000 white papers, case studies and analyst reports.
Do you agree?
confused me with all other reasons..
After I read your article, I found another one on ZDNet. Russell Shaw with ZDNet was saying another reason(seems like different from yours). Should I use this patch or not wit skype..? http://blogs.zdnet.com/ip-telephony/?p=2208 I blog your article on mine
Posted by: Song Kim 20 Aug 2007
No accountability
That reminds me of health clubs that over sell membership cause they know that a large % of the members will never show up then blaming highway construction when their clubs are now overcrowded. "Normally, a "self-healing" feature would intervene and allow the system to correct or compensate for the increased load. That feature, however, never activated and the result was a chain reaction that took down the entire service. " Massive simulteneous user logins should never have broken it in the first place, nevermind the "self healing" instead, anything over what was "healthy" should have been denied/postponed until the servers could catch up. Then to pass the blame onto the creator of the operating system that most of their users use because the users rebooted their machines. Weak. I'm no Fan of Microsloth but this attack on them due to Skypes incompetence is pathetic.
Posted by: Michael Lang 20 Aug 2007