Print

Print


Hi Jerry,
no this has nothing to do with the load.
From LCG1-1_0_0 on we introduced a mechanism that
avoids some of the problems that we had seen with the GASS Cache and
NFS (especially with PBS). This system does updates at a low frequency
which results for small test jobs in a noticable increase in latency.
90sec are what you can expect (for long jobs this doesn't matter, for
testing it is a bit painful)

        markus


On Fri, 7 Nov 2003, Jerry Gieraltowski wrote:

> Markus - How long did it take to get a response?  I just submitted your exact
> command from a server at the University of Chicago (running VDT-1.1.11) and it
> did respond after 90 seconds.  Seems a bit slow to me.  Maybe the system is
> overloaded?
> Jerry
>
>
> Markus SCHULZ wrote:
>
> > No,
> > I just tried it:
> > [adc0014] ~/> globus-job-run adc0015.cern.ch/jobmanager-lcgpbs /bin/pwd
> > /home/dteam007/globus-tmp.lxshare0227.12354.0
> >
> >   markus
> >
> > On Thu, 6 Nov 2003, Andrea Sciaba' wrote:
> >
> > >
> > >
> > > -------- Original Message --------
> > > Subject: No response from adc0015.cern.ch/jobmanager-lcgpbs
> > > Date: Thu, 06 Nov 2003 14:17:35 -0600
> > > From: Jerry Gieraltowski <[log in to unmask]>
> > > Reply-To: Jerry Gieraltowski <[log in to unmask]>
> > > To: LCG-1 Support <[log in to unmask]>
> > > CC: Flavia Donno <[log in to unmask]>, Rob Gardner <[log in to unmask]>
> > >
> > > pbs jobmanager stopped responding around 13:00CST (USA)
> > > Any idea why the queue stopped?
> > > Jerry Gieraltowski
> > > Argonne National Laboratory
> > >
> > >
> > >
> >
> > --
> > *************************************************************************
> > *                                                                       *
> > *  CERN                                    Markus W. Schulz             *
> > *  Bat. 31 2-015                                                        *
> > *  CH-1211 Geneva 23                                                    *
> > *                                                                       *
> > *  Phone: +41 22 76 77909                                               *
> > *  www.cern.ch                                                          *
> > *                                                                       *
> > *************************************************************************
>
>

--
*************************************************************************
*                                                                       *
*  CERN                                    Markus W. Schulz             *
*  Bat. 31 2-015                                                        *
*  CH-1211 Geneva 23                                                    *
*                                                                       *
*  Phone: +41 22 76 77909                                               *
*  www.cern.ch                                                          *
*                                                                       *
*************************************************************************