Ian
It was certainly slowed to the point of intermittantly accepting new
jobs and occasionally responding to BDII queries. The fork jobs were
using a lot of CPU and memory was swapped to the maximum. It looked like
there were more jobs than on per user per VO.
cheers
Steve
> -----Original Message-----
> From: Testbed Support for GridPP member institutes
> [mailto:[log in to unmask]] On Behalf Of Ian Stokes-Rees
> Sent: 11 August 2006 16:06
> To: [log in to unmask]
> Subject: Re: shared experiment area load
>
> I'm curious as to how 4000 jobs on a CE can kill it. Surely
> the CE for a large cluster would be expected to handle 10,000
> or more jobs.
>
> Cheers,
>
> Ian
>
> David McBride wrote:
> > On Fri, 2006-08-11 at 13:34 +0100, Steve Traylen wrote:
> >
> >> I doubt that "they" are sending fork jobs the RB does send
> fork jobs
> >> as part of normall operation. One per user per RB.
> >
> > I'm somewhat dismayed that this (security!) bug has still not been
> > properly fixed yet...
> >
> > David
>
>
> --
> Ian Stokes-Rees [log in to unmask]
> Particle Physics, Oxford http://grid.physics.ox.ac.uk/~stokes
>
|