AC2 rollbacks avoided...
Posted on Monday, January 27 @ 18:54:12 PST by kadams
Microsoft and Turbine have made an official proclamation regarding the recent virus-bedevilled AC2 server situation. But I'm left wondering...
The proclamation says clearly that Microsoft and Turbine had patched their servers appropriately so that they were not impacted directly by the virus. Fair enough: I'm glad to hear that. But the side effects of the worm which, presumably, should have been restricted to bandwidth congestion seem to have caused an amazing number serious problems:
their servers lost track of inventory and character information
portals and chests in dungeons failed to work
.Net passport (login) services failed
their administration tools failed totally, to the point where they couldn't even update their status web page
I'm a fairly supportive guy. I'll assume that Microsoft and Turbine had a bunch of code that had never been stressed by network connectivity problems before. But this kind of thing can happen even without a worm or virus (E.G.: when several pieces of network infrastructure fail simultaneously). I would propose that Microsoft and Turbine had better take what happened as a learning experience and improve the robustness of their systems.
They should add some code on the servers to deal more efficiently with rapid and significant connection loss. They had better set up a locally attached terminal connected to their annoucement servers (which, oddly enough, I was able to get to to *read* all day all the way from Canada, without any connection problems...yet they couldn't update them). And maybe they should double-check their procedures so that their operational staff don't make situations like this worse next time by rebooting systems and causing data loss.
