1 | $ sudo /etc/init.d/slurm-llnl start |
1 2 3 4 | $ sinfo PARTITION AVAIL TIMELIMIT NODES STATE NODELIST debug* up infinite 1 idle mtj-VirtualBox $ |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | $ scontrol show partition PartitionName=debug AllocNodes=ALL AllowGroups=ALL Default=YES DefaultTime=NONE DisableRootJobs=NO Hidden=NO MaxNodes=UNLIMITED MaxTime=UNLIMITED MinNodes=1 Nodes=mtj-VirtualBox Priority=1 RootOnly=NO Shared=NO PreemptMode=OFF State=UP TotalCPUs=1 TotalNodes=1 $ scontrol show node mtj-VirtualBox NodeName=mtj-VirtualBox Arch=i686 CoresPerSocket=1 CPUAlloc=0 CPUErr=0 CPUTot=1 Features=(null) Gres=(null) OS=Linux RealMemory=1 Sockets=1 State=IDLE ThreadsPerCore=1 TmpDisk=0 Weight=1 BootTime=2012-03-07T14:59:01 SlurmdStartTime=2012-04-17T11:10:43 Reason=(null) |
1 2 3 4 5 6 7 8 9 10 | $ srun -l hostname 0: mtj-VirtualBox $ srun -l sleep 5 & [1] 24127 $ squeue JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON) 15 debug sleep mtj R 0:03 1 mtj-VirtualBox $ [1]+ Done srun -l sleep 5 $ |
1 2 3 4 5 6 7 8 9 10 11 12 13 | $ srun -l sleep 60 & [1] 24262 $ squeue JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON) 16 debug sleep mtj R 0:03 1 mtj-VirtualBox $ scancel 16 srun: Force Terminated job 16 $ srun: Job step aborted: Waiting up to 2 seconds for job step to finish. 0: slurmd[mtj-VirtualBox]: error: *** STEP 16.0 CANCELLED AT 2012-04-17T12:08:08 *** srun: error: mtj-VirtualBox: task 0: Terminated [1]+ Exit 15 srun -l sleep 60 $ |
1 2 3 4 5 | $ sudo /etc/init.d/slurm-llnl stop * Stopping slurm central management daemon slurmctld [ OK ] * Stopping slurm compute node daemon slurmd [ OK ] slurmd is stopped $ |
欢迎光临 电子技术论坛_中国专业的电子工程师学习交流社区-中电网技术论坛 (http://bbs.eccn.com/) | Powered by Discuz! 7.0.0 |