 |
 |
> On 4 Jun 2016, at 12:05, sjones <[log in to unmask]> wrote: > > We in Liverpool are in the same boat, Peter. > > On 2016-06-03 15:01, Peter Gronbech wrote: >> Hi Andrew, >> Is this still the case, our VAC nodes are also mostly empty but our >> condor WN's behind the ARC ce are all happily running jobs. I don’t think anything’s broken. It’s the pattern of what job types have been running and where they’re enabled. This is what we’ve seen at your LCG (i.e. ARC) and VAC sites: That's the job completion rate, so the number of jobs running can be an order of magnitude higher. What happens after the 1st of June is that the number of Monte Carlo jobs runs down but the user jobs (here just for Liverpool) pick up: At the moment, the Vac sites aren’t set up in LHCb DIRAC to get user jobs, just MC. So when there’s not much MC available, we get a quiet period. The LCG (ARC) sites have been enabled for user jobs. I’ll look at making this consistent across sites as we now have that Tier-2A policy I talked about in Pitlochry. Btw, a quick way of seeing what is going on across the LHCb Vac/Vcycle sites is to look at Ganglia: http://lgm.cern.ch/?r=day&cs=&ce=&m=load_one&s=by+name&c=&tab=m&vn=&hide-hf=false Cheers Andrew >> Pete >> -- >> ---------------------------------------------------------------------- >> Peter Gronbech GridPP Project Manager Tel No. : 01865 273389 >> Department of Particle Physics, >> University of Oxford, >> Keble Road, Oxford OX1 3RH, UK E-mail : [log in to unmask] >> ---------------------------------------------------------------------- >>> -----Original Message----- >>> From: Testbed Support for GridPP member institutes [mailto:TB- >>> [log in to unmask]] On Behalf Of Andrew McNab >>> Sent: 25 May 2016 12:33 >>> To: [log in to unmask] >>> Subject: Re: LHCB Dirac VAC Jobs >>> > On 25 May 2016, at 12:19, sjones <[log in to unmask]> wrote: >>> > >>> > Just noting: our LHCB Dirac (i.e. VAC) Jobs have dried up: >>> > >>> > See photo attached: >>> > >>> > Must be a job drought on? >>> Yes, LHCb MC production has trailed off across all sites today. >>> Cheers >>> Andrew Cheers Andrew -- Dr Andrew McNab University of Manchester High Energy Physics, LHCb@CERN (Distributed Computing Coordinator), and GridPP (LHCb + Tier-2 Evolution) www.hep.manchester.ac.uk/u/mcnab Skype: andrew.mcnab.uk