|
Rattled Skype Tells It Twice |
|
|
|
Written by Adam Gosling
|
|
Wednesday, 22 August 2007 |
Besieged
by a world of users that don't believe its explanation for the two day
outage in its peer-to-peer VoIP services, Skype has tried again to
explain what happened to its systems, though its expanded explanation
is unlikely to appease the more suspicious commentators.
Also,
apparently responding to criticism that eBay and Skype senior
management ducked for cover throughout the entire episode, Skype's home
page now has a message from Skype CEO and co-founder Niklas Zennström.
And just in case people don't believe him either, they've taken to
repeating themselves immediately instead of waiting for the perfect
storm of criticism to gather strength.
When
VoIP News checked back at Skype HQ in the World Wide Web today,
Zennström's personal apology (well it's not really an apology) is
pasted not once but twice into the web page - see pic below. When
things go wrong they just keep going wrong.
Zennström's message to Skype users is actually a thankyou
to the readers who "contacted us with messages of support" rather than
to the disgruntled subscribers who were criticising them. Apparently
its the former, rather than the latter who "make the Skype community
what it is today".
Meanwhile, Skype blogger Villu Arak
has had another crack at explaining why the company's services to
customers was inoperable for two days. After yesterday's explanation
raised more than a few eyebrows and sent journalists scurrying to
disprove the company's claims, Skype has been forced to dole out a
little more detail.
While clinging to the claim that
Microsoft's Patch Tuesday reboot cycle prompted the outage, Skype
emphatically tries to say that its not "blaming Microsoft for what
happened". And stresses that "there was nothing different about this
set of Microsoft
patches" which had already been established by journalists checking
with Microsoft.
"We don't blame anyone but ourselves," writes Arak explaining the patches
were merely a trigger for what he describes as a "perfect storm" of events that led to
the disruption.
Previous patch cycle reboots didn't cause the outage "because the update patches were not the cause of the
disruption. In previous instances where a large number of supernodes in
the P2P network were rebooted, other factors of a "perfect storm" had
not been present. That is, there had not been such a combination of
high usage load during supernode rebooting. As a result, P2P network
resources were allocated efficiently and self-healing worked fast
enough to overcome the challenge," he says.
In short what the company says again, in a little more detail is
that sufficient supernodes rebooted simultaneously, while at the same
time there was a significant P2P
network load (we presume they mean users talking or trying to log in
after a reboot) to bring down the entire house of cards.
Arak's more detailed explanation also reiterates that "a previously unseen fault in the P2P
network resource allocation algorithm" meant the P2P
network was not able to self-heal.
With the 'core' network taken out the system "needed outside
intervention and assistance" from Skype's engineers to get it back up and running says Arak.
"Once we found the algorithmic fix to ensure continued
operation in the face of high numbers of client reboots, the efforts
focused squarely on stabilising the P2P core," he continues adding that the P2P core can now cope with simultaneous P2P
network load and core size changes similar to those that occurred on
August 16.
  
Related news items Newer news items
Older news items
|
|
|
|
