Re:Skype Outage – I run a supernode and here’s what I think ...
by Groxx on Friday December 24, 2010 @03:41AM
(#2036895)
A wonderful reason to have pre-release testers for every piece of software that communicates with other software, any time you plan to push updates. With pulls, your early updaters are your test beds; with push, you have to create your testing groups.
[ Parent
| on HN ]
Re:Skype Outage – I run a supernode and here’s what I think ...
by guelo on Friday December 24, 2010 @04:21AM
(#2036949)
The problem with a big adaptive p2p network is that it is close to impossible to test under real production loads.
Re:Skype Outage – I run a supernode and here’s what I think ...
by jrockway on Friday December 24, 2010 @07:01AM
(#2037135)
My guess is that the average supernode is on something like 6M/1.5M DSL. This is easy to simulate.
Now, if most of your nodes are on machines directly connected to 10GbE, then that's a problem. But most Skype nodes aren't. (I do imagine there are a few connected to 100M+ connections. But only a few.)
Re:Skype Outage – I run a supernode and here’s what I think ...
by ohashi on Friday December 24, 2010 @12:27PM
(#2037597)
I imagine quite a lot in countries like Sweden. I had 100Mbit in my apartment for cheaper than my shitty comcast which barely breaks 5.
Re:Skype Outage – I run a supernode and here’s what I think ...
by pyre on Friday December 24, 2010 @12:33PM
(#2037619)
They could have setup their 'mega-supernodes' running the new client in a production environment, so that if it crashed horribly, it wasn't a client machine that failed (or their entire network).
Upgrading (relatively) untested software network-wide on what is essentially their critical infrastructure is bad news. If this didn't happen now, it would happen sometime. It was just a question of when.
Re:Skype Outage – I run a supernode and here’s what I think ...
by Daniel_Newby on Friday December 24, 2010 @04:14AM
(#2036939)
Push software is even better about testing: you can forcibly upgrade 1% of installs, monitor them intensely, and let the (non-upgraded) system executive roll back the version if something bad happens. The system executive can even automatically roll back if it loses contact with the vendor, in case the upgrade goes so horribly wrong that it brings down networking.