> -----Original Message-----
> From: Testbed Support for GridPP member institutes [mailto:TB-
>
> On 14 Oct 2010, at 12:17, John Gordon wrote:
>
> > Forwarded from Gilles.
> >
> >> We can't put whole sites in downtime anymore, this has to been done
> at service level.
>
> That strikes me as a regression. Putting the whole site down is mostly
> done in extremis, when there's a power cut or network outage.
>
> I can't fathom the rationale for removing that feature, given that it
> will only result in sucking up a lot of admin time, at a point when
> it's a scarce commodity. And result in errors thoroughly messing up
> stats, if one service is missed.
>
Hear, hear.
This certainly seems to be a clear change for the worse. What is
the reasoning behind it?
Ewan
|