> -----Original Message-----
> From: Testbed Support for GridPP member institutes [mailto:TB-
> [log in to unmask]] On Behalf Of Sam Skipsey
>
> It would possibly be good to bring this up in the monitoring / infosys
> session at WLCG Workshop this afternoon? As an object lesson...
>
But what is the lesson? Is it:
- Don't put things in full production without testing them,
- Sudden simultaneous site failures for something that barely ever changes are probably not site failures so don't ticket them,
- The real test is real work, so don't worry about stuff you're not using,
Or is it, indeed, that we've built a system of both technology and communication so robust that someone can screw up with a bit of central international infrastructure, nothing breaks, and it all gets put back within a few days (including a weekend in the middle), and that's pretty good?
Ewan
|