What about the pbs_mom logs on one the workers.
Steve
On Thu, 13 Mar 2003, D.Kant wrote:
> Hmm...
>
> Log files show this:
>
> 03/13/2003 11:46:51;0002; pbs_sched;Svr;Log;Log opened
> 03/13/2003 11:46:51;0002; pbs_sched;Svr;die;caught signal 15
> 03/13/2003 11:46:51;0002; pbs_sched;Svr;Log;Log closed
> 03/13/2003 11:46:51;0002; pbs_sched;Svr;Log;Log opened
> 03/13/2003 11:46:51;0002; pbs_sched;Svr;main;/usr/pbs/sbin/pbs_sched startup pid 10692
> 03/13/2003 11:46:52;0080; pbs_sched;Svr;main;brk point 134688768
>
> Dave.
>
>
> On Thu, 13 Mar 2003, Frederic Brochu wrote:
>
> > Hello,
> >
> > You have certainly a problem with PBS: 6 jobs are queued and none running:
> >
> > [brochu@gppui04 scripts]$ globus-job-run hepbf4.ph.qmul.ac.uk
> > /usr/pbs/bin/qstat -a
> >
> > hepbf4.ph.qmul.ac.uk:
> > Req'd Req'd
> > Elap
> > Job ID Username Queue Jobname SessID NDS TSK Memory Time S
> > Time
> > --------------- -------- -------- ---------- ------ --- --- ------ ----- -
> > -----
> > 0.hepbf4.ph.qmu iteam001 S STDIN -- -- 1 -- 30:00 Q
> > --
> > 1.hepbf4.ph.qmu cms001 M STDIN -- -- 1 -- 30:00 Q
> > --
> > 2.hepbf4.ph.qmu cms001 L STDIN -- -- 1 -- 30:00 Q
> > --
> > 3.hepbf4.ph.qmu cms001 S STDIN -- -- 1 -- 30:00 Q
> > --
> > 4.hepbf4.ph.qmu iteam001 S STDIN -- -- 1 -- 30:00 Q
> > --
> > 5.hepbf4.ph.qmu iteam001 L STDIN -- -- 1 -- 30:00 Q
> > --
> >
> > Best regards,
> > Frederic
> >
> > On Thu, 13 Mar 2003, D.Kant wrote:
> >
> > > Hi Everyone,
> > >
> > > I still don't see why I'm not green.
> > > Both hepbf2 and hepbf4 are reporting information which is accessible via IC RB
> > > http://www.hep.ph.ic.ac.uk/~dguser/diagnostics.html
> > >
> > > Details are my setup are here:
> > > http://hepwww.ph.qmul.ac.uk/~kant/post-install.html
> > >
> > > Perhaps there are some "misconfigurations" in my site-cfg.h such as
> > > the TOP GIIS?
> > >
> > > If a potential firewall problem, then what ports MUST be open??
> > >
> > > My CE reports that both itself and the WN are ready to process jobs:
> > > "/usr/pbs/bin/pbsnodes -a"
> > > hepbf4.ph.qmul.ac.uk
> > > state = free
> > > np = 2
> > > ntype = cluster
> > >
> > > hepbf5.ph.qmul.ac.uk
> > > state = free
> > > np = 2
> > > ntype = cluster
> > >
> > >
> > > Dave.
> > >
> > >
> > > >Date: Tue, 11 Mar 2003 21:18:29 +0000 (GMT)
> > > >From: Andrew McNab <[log in to unmask]>
> > > >To: D.Kant <[log in to unmask]>
> > > >Subject: Re: New map page and amber status
> > > >
> > > >On Tue, 11 Mar 2003, D.Kant wrote:
> > > >
> > > >>
> > > >> This is good...but clearly i'm a bit puzzled why we don't have
> > > >> a green :-(
> > > >>
> > > >> What things must be "in place" in order to get communication via the
> > > >> Gridpp RB?
> > >
> > > >Looks like I can't get a job submission to you to get past the Scheduled
> > > >state. This may well be a firewall problem (EDG uses Globus's Two Phase
> > > >Commit job submission and the firewall requirements for this are even
> > > >worse than for default globus-job-submit.) It's worth asking on the
> > > >tb-support list to get Steve Traylen and Dave Colling's advice (and other
> > > >people) advice about getting thing working via the RB.
> > > >
> > > >I think you are very close to getting it working btw.
> > > >
> > > >Cheers,
> > > >
> > > > Andrew
> > >
> >
>
> --
> --------------------------------------------------------------
> Department of Physics | Dr Dave Kant
> Queen Mary College | TEL/FaX: +44 (0)20 7882 5054
> Mile End Road London E1 4NS | e-mail : [log in to unmask]
> --------------------------------------------------------------
>
--
Steve Traylen
[log in to unmask]
http://www.gridpp.ac.uk/
|