1SJSTAT(1) User Contributed Perl Documentation SJSTAT(1)
2
6 sjstat - List attributes of jobs under the SLURM control
7
9 sjstat [-h ] [-c] [-r ] [-v]
10
12 The sjstat command is used to display statistics of jobs under control
13 of SLURM. The output is designed to give information on the resource
14 usage and availablilty, as well as information about jobs that are
15 currently active on the machine. This output is built using the SLURM
16 utilities, sinfo, squeue and scontrol, the man pages for these utilites
17 will provide more information and greater depth of understanding.
18
20 -h Display a brief help message
21 -c Display the computing resource information only.
23 -man
25 Show the man page.
26 -r Display only the running jobs.
28 -v Display more verbose information.
30
32 The following is a basic request for status.
33 > sjstat
35 Scheduling pool data:
37 ------------------------------------------------------------
38 Pool Memory Cpus Total Usable Free Other Traits
39 ------------------------------------------------------------
40 pdebug 15000Mb 8 32 32 24 (null)
41 pbatch* 15000Mb 8 1072 1070 174 (null)
42 Running job data:
45 -------------------------------------------------------------------
46 JobID User Nodes Pool Status Used Master/Other
47 -------------------------------------------------------------------
48 395 mary 1000 pbatch PD 0:00 (JobHeld)
49 396 mary 1000 pbatch PD 0:00 (JobHeld)
50 375 sam 1000 pbatch CG 0:00 (JobHeld)
51 388 fred 32 pbatch R 25:27 atlas89
52 361 harry 512 pbatch R 1:01:12 atlas618
53 1077742 sally 8 pdebug R 20:16 atlas18
54 The Scheduling data contains information pertaining to the:
57 Pool a set of nodes
59 Memory the amount of memory on each node
60 Cpus the number of cpus on each node
61 Total the total number of nodes in the pool
62 Usable total usaable nodes in the pool
63 Free total nodes that are currently free
64 The Running job data contains information pertaining to the:
66 JobID the SLURM job id
68 User owner of the job
69 Nodes nodes required, or in use by the job
70 (Note: On cpu scheduled machines, this field
71 will be labled "Procs" show the number of processors
72 the job is using.)
73 Pool the Pool required or in use by the job
74 Status current status of the job
75 Used Wallclick time used by the job
76 Master/Other Either the Master (head) node used by the job, or may
77 indicate furhter status of a pending, or completing job.
78 The common status values are:
80 R The job is running
82 PD The job is Pending
83 CG The job is Completing
84 These are states reproted by SLURM and more elaborate docuemntation
86 can be found in the squeue/sinfo man pages.
87 An example of the -v option.
90 Scheduling pool data:
92 -----------------------------------------------------------------------------
93 Total Usable Free Node Time Other
94 Pool Memory Cpus Nodes Nodes Nodes Limit Limit Traits
95 -----------------------------------------------------------------------------
96 pdebug 15000Mb 8 32 32 24 16 30 (null)
97 pbatch* 15000Mb 8 1072 1070 174 UNLIM UNLIM (null)
98 Running job data:
100 ---------------------------------------------------------------------------------------------------
101 Time Time Time
102 JobID User Nodes Pool Status Used Limit Started Master/Other
103 ---------------------------------------------------------------------------------------------------
104 38562 tom 4 pbatch PD 0:00 1:00:00 01-14T18:11:22 (JobHeld)
105 The added fields to the "Scheduling pool data" are:
107 Node Limit SLURM imposed node limit.
109 Time Limit SLURM imposed time limit, value in minutes.
110 The added fields to the "Running job data" are:
112 Limit Time limit of job.
114 Start Start time of job.
115
117 Report bugs to <eckert2@llnl.gov>
118perl v5.30.1 2020-01-30 SJSTAT(1)