Skype

[bascule]

http://techdirt.com/articles/20070820/163914.shtml

NO SHIT

Anyone else who doesn't think the "BLAME MICROSOFT" story holds water?

[theprez98]

Re: Skype

http://techdirt.com/articles/20070820/163914.shtml

NO SHIT

Anyone else who doesn't think the "BLAME MICROSOFT" story holds water?

Normally Skype’s peer-to-peer network has an inbuilt ability to self-heal, however, this event revealed a previously unseen software bug within the network resource allocation algorithm which prevented the self-healing function from working quickly.

I don't read it as Skype blaming Microsoft, the "official" Skype explanation suggests that the Windows Update reboots triggered an "unseen software bug". So the way I read it, they seem to be taking responsibility, at least on some level.

[Thorn]

Re: Skype

This past winter, a client's SBS2003 server had an issue after an MS security update and reboot. As it turned out, the DNS entries where corrupted, and in looking at some files timestamps it looked like it was around the time of the reboot. Was it MS's or someone else's fault? I don't know, but the client had to deal with the server not dealing correctly with connections until it was discovered and fixed.

[HighWiz]

Re: Skype

That's the problem when anything is set to "Automatic Update".

[Thorn]

Re: Skype

That's the problem when anything is set to "Automatic Update".

A very good point. Although most clients don't want to pay me for scheduled maintenance for something they see as being "free" from MS. In the end, they pay me later (and more) to fix it.

[noid]

Re: Skype

I dont like automatic updates on servers. You never know when some application or bit of code on your production server is going to choke on a patch (and this is OS agnostic, I've seen Solaris/Linux boxes choke up on patches too). I recommend testing the patch on a lab machine for at least 24 hours before deployment. Just make sure your lab machine is a mirror copy of the production server.