On Thu, 20 Mar 2003, Frederic Brochu wrote:
> Hi,
>
> This is symptomatic of quotas problems for /home in farm003. (we have
> installed group quotas on farm003, but they still need some tuning.)
> Santanu, Chris, can you check quotas on the iteam group ?
>
> P.S: serv01 is not yet configured as a full fledge SE: it is "online" as
> its various RAID components can be NFS mounted from farm020, but that's
> all of the story.
Can you arrange for serv01 not to register into any giis for the moment
or cut its grid-mapfile down to your local users.
Steve
>
> P.P.S: I thought demos would go under the tutor VO?
>
> Best regards,
> Frederic
>
> On Thu, 20 Mar 2003, Stephen Burke wrote:
>
> > Hi,
> >
> > There seems to be some kind of problem at Cambridge:
> >
> > globus-job-run farm003.hep.phy.cam.ac.uk /bin/pwd
> > GRAM Job submission failed because the gatekeeper failed to run the job
> > manager (error code 47)
> >
> > globus-job-run farm020.hep.phy.cam.ac.uk /bin/pwd
> > GRAM Job submission failed because the job manager failed to open stdout
> > (error code 73)
> >
> > I saw the same kind of thing when I was doing a demo last week, jobs
> > would be sent to Cambridge and immediately fail. This seems to be
> > something VO-related; I get the above with my iteam certificate,
> > but with Atlas it works:
> >
> > globus-job-run farm003.hep.phy.cam.ac.uk /bin/pwd
> > /home/atlas014
> >
> > globus-job-run farm020.hep.phy.cam.ac.uk /bin/pwd
> > /home/atlas014
> >
> > Both certificates appear in the list of authorised users.
> >
> > Stephen
> >
>
--
Steve Traylen
[log in to unmask]
http://www.gridpp.ac.uk/
|