[slurm-dev] Re: Failed to allocate resource : Unable to contact slurm controller
Umm... I don't have any file named slurm or slurm-llnl in my /etc/init.d On Tue, Sep 17, 2013 at 4:46 PM, Sivasangari Nandy < sivasangari.na...@irisa.fr> wrote: > hello, > > try this : > /etc/init.d/slurm-llnl start > /etc/init.d/slurm-llnl stop > /etc/init.d/slurm-llnl startclean > > > > > -- > > *De: *"Arjun J Rao" > *À: *"slurm-dev" > *Envoyé: *Mardi 17 Septembre 2013 11:33:53 > *Objet: *[slurm-dev] Failed to allocate resource : Unable to contact > slurm controller > > > I want to run SLURM on on a single machine as a proof of concept to run > some trivial MPI programs on my machine. > I keep getting the message : > Failed to allocate resources ; Unable to contact > slurm controller > In my slurm.conf file, I have named the ControlMachine as localhost and > ControlAddr as 127.0.0.1 > Compute Node name as localhost and NodeAddr as 127.0.0.1 too. > > What am i doing wrong ? > > Scientific Linux 6.4 > 64-bit > > > > > -- > *Siva*sangari NANDY - Plate-forme *GenOuest* > IRISA-INRIA, Campus de Beaulieu > 263 Avenue du Général Leclerc > 35042 Rennes cedex, France > Tél: +33 (0) 2 99 84 25 69 > Bureau : D152 > >
[slurm-dev] Re: Failed to allocate resource : Unable to contact slurm controller
hello, try this : /etc/init.d/slurm-llnl start /etc/init.d/slurm-llnl stop /etc/init.d/slurm-llnl startclean - Mail original - > De: "Arjun J Rao" > À: "slurm-dev" > Envoyé: Mardi 17 Septembre 2013 11:33:53 > Objet: [slurm-dev] Failed to allocate resource : Unable to contact > slurm controller > Failed to allocate resource : Unable to contact slurm controller > I want to run SLURM on on a single machine as a proof of concept to > run some trivial MPI programs on my machine. > I keep getting the message : > Failed to allocate resources ; Unable to contact slurm controller > In my slurm.conf file, I have named the ControlMachine as localhost > and ControlAddr as 127.0.0.1 > Compute Node name as localhost and NodeAddr as 127.0.0.1 too. > What am i doing wrong ? > Scientific Linux 6.4 > 64-bit -- Siva sangari NANDY - Plate-forme GenOuest IRISA-INRIA, Campus de Beaulieu 263 Avenue du Général Leclerc 35042 Rennes cedex, France Tél: +33 (0) 2 99 84 25 69 Bureau : D152
[slurm-dev] Failed to allocate resource : Unable to contact slurm controller
I want to run SLURM on on a single machine as a proof of concept to run some trivial MPI programs on my machine. I keep getting the message : Failed to allocate resources ; Unable to contact slurm controller In my slurm.conf file, I have named the ControlMachine as localhost and ControlAddr as 127.0.0.1 Compute Node name as localhost and NodeAddr as 127.0.0.1 too. What am i doing wrong ? Scientific Linux 6.4 64-bit
[slurm-dev] mpich2 to use multiple machines
Hello, I got a small problem with mpich2 for slurm. I want to run my jobs in more than one machines (here for the test I just wanted it with VM-669, so I've put in the file Mname.txt just VM-669) From the master (VM-667) I run : mpiexec -machinefile Mname.txt -np 1 /bin/spleep 60 but I have these errors : [proxy:0:0@VM-669] launch_procs (./pm/pmiserv/pmip_cb.c:687): unable to change wdir to /root/omaha-beach (No such file or directory) [proxy:0:0@VM-669] HYD_pmcd_pmip_control_cmd_cb (./pm/pmiserv/pmip_cb.c:935): launch_procs returned error [proxy:0:0@VM-669] HYDT_dmxu_poll_wait_for_event (./tools/demux/demux_poll.c:77): callback returned error status [proxy:0:0@VM-669] main (./pm/pmiserv/pmip.c:226): demux engine error waiting for event [mpiexec@VM-667] control_cb (./pm/pmiserv/pmiserv_cb.c:215): assert (!closed) failed [mpiexec@VM-667] HYDT_dmxu_poll_wait_for_event (./tools/demux/demux_poll.c:77): callback returned error status [mpiexec@VM-667] HYD_pmci_wait_for_completion (./pm/pmiserv/pmiserv_pmci.c:181): error waiting for event [mpiexec@VM-667] main (./ui/mpich/mpiexec.c:405): process manager error waiting for completion Have you got any idea ? advance thanks, Siva -- Siva sangari NANDY - Plate-forme GenOuest IRISA-INRIA, Campus de Beaulieu 263 Avenue du Général Leclerc 35042 Rennes cedex, France Tél: +33 (0) 2 99 84 25 69 Bureau : D152
[slurm-dev] Limiting number of CPUs per Job
Hi, I want to have a partition for serial jobs: Each job running with only a single CPU, many jobs on a single node. How would I do that? MaxCPUs or QOS seem to relate to associations (and not to single partitions) or could I use that somehow? Regards, Olaf -- Dipl. Inform. Olaf Gellertemail gell...@dkrz.de Deutsches Klimarechenzentrum GmbH phone +49 (0)40 460094 214 Bundesstrasse 45a fax+49 (0)40 460094 270 D-20146 Hamburg, Germany wwwhttp://www.dkrz.de Sitz der Gesellschaft: Hamburg Gesch�ftsf�hrer: Prof. Dr. Thomas Ludwig Registergericht: Amtsgericht Hamburg, HRB 39784