Erica Riello
2014-09-10 12:21:34 UTC
Hi all,
I have a 3 nodes cluster running Slurm 14.03.07 and I've been experiencing
some problems with submission.
Depending on the directory a job is submitted, it is not executed:
Example 1: working
[torquepbs:~/teste_slurm] pwd
/home/msv/ericaflr/teste_slurm
[torquepbs:~/teste_slurm] more echoandsleepjob.sh
#!/bin/sh
echo 'Hello from Slurm'
sleep 30
[torquepbs:~/teste_slurm] sbatch echoandsleepjob.sh
Submitted batch job 88
[torquepbs:~/teste_slurm] squeue
JOBID PARTITION NAME USER ST TIME NODES
NODELIST(REASON)
88 particao1 echoands ericaflr R 0:01 1
torquepbsno1
[torquepbs:~/teste_slurm] more slurm-88.out
Hello from Slurm
Example 2: not working
[torquepbs:/tmp/teste_slurm] pwd
/tmp/teste_slurm
[torquepbs:/tmp/teste_slurm] more echoandsleepjob.sh
#!/bin/sh
echo 'Hello from Slurm'
sleep 30
[torquepbs:/tmp/teste_slurm] sbatch echoandsleepjob.sh
Submitted batch job 87
[torquepbs:/tmp/teste_slurm] squeu
squeu: Command not found.
[torquepbs:/tmp/teste_slurm] ls
destinreqjob2.sh* destinreqjob.sh* directoryjob.sh* echoandsleepjob.sh*
echojob.sh* input.txt slurm.conf stageinjob.sh*
As I explained, it doesn't show the job executing in the partition. What
might be wrong?
Thanks in advance.
Regards,
I have a 3 nodes cluster running Slurm 14.03.07 and I've been experiencing
some problems with submission.
Depending on the directory a job is submitted, it is not executed:
Example 1: working
[torquepbs:~/teste_slurm] pwd
/home/msv/ericaflr/teste_slurm
[torquepbs:~/teste_slurm] more echoandsleepjob.sh
#!/bin/sh
echo 'Hello from Slurm'
sleep 30
[torquepbs:~/teste_slurm] sbatch echoandsleepjob.sh
Submitted batch job 88
[torquepbs:~/teste_slurm] squeue
JOBID PARTITION NAME USER ST TIME NODES
NODELIST(REASON)
88 particao1 echoands ericaflr R 0:01 1
torquepbsno1
[torquepbs:~/teste_slurm] more slurm-88.out
Hello from Slurm
Example 2: not working
[torquepbs:/tmp/teste_slurm] pwd
/tmp/teste_slurm
[torquepbs:/tmp/teste_slurm] more echoandsleepjob.sh
#!/bin/sh
echo 'Hello from Slurm'
sleep 30
[torquepbs:/tmp/teste_slurm] sbatch echoandsleepjob.sh
Submitted batch job 87
[torquepbs:/tmp/teste_slurm] squeu
squeu: Command not found.
[torquepbs:/tmp/teste_slurm] ls
destinreqjob2.sh* destinreqjob.sh* directoryjob.sh* echoandsleepjob.sh*
echojob.sh* input.txt slurm.conf stageinjob.sh*
As I explained, it doesn't show the job executing in the partition. What
might be wrong?
Thanks in advance.
Regards,
--
===============
Erica Riello
Aluna Engenharia de Computação PUC-Rio
===============
Erica Riello
Aluna Engenharia de Computação PUC-Rio