1sstat(1)                        Slurm Commands                        sstat(1)
2
3
4

NAME

6       sstat - Display various status information of a running job/step.
7
8

SYNOPSIS

10       sstat [OPTIONS...]
11
12

DESCRIPTION

14       Status information for running jobs invoked with Slurm.
15
16       The  sstat  command  displays job status information for your analysis.
17       The sstat command displays information pertaining to CPU,  Task,  Node,
18       Resident  Set  Size  (RSS) and Virtual Memory (VM).  You can tailor the
19       output with the use of the --fields= option to specify the fields to be
20       shown.
21
22       For  the  root user, the sstat command displays job status data for any
23       job running on the system.
24
25       For the non-root user, the sstat output is limited to the user's jobs.
26
27
28       NOTE:  The sstat command requires that the jobacct_gather plugin be in‐
29       stalled and operational.
30
31       NOTE:  The sstat command is not supported on Cray ALPS.
32
33
34       -a, --allsteps
35              Print all steps for the given job(s) when no step is specified.
36
37
38       -e, --helpformat
39              Print a list of fields that can be specified with the '--format'
40              option.
41
42
43       -h, --help
44              Displays a general help message.
45
46
47       -i, --pidformat
48              Predefined format to list the pids running for  each  job  step.
49              (JobId,Nodes,Pids)
50
51
52       -j, --jobs
53              Format  is  <job(.step)>.  Stat this job step or comma-separated
54              list of job steps. This option is required.   The  step  portion
55              will  default  to  the  lowest numbered (not batch, extern, etc)
56              step running if not specified, unless the --allsteps flag is set
57              where  not specifying a step will result in all running steps to
58              be displayed.  NOTE: A step id of 'batch' will display  the  in‐
59              formation  about  the  batch  step.  NOTE: A step id of 'extern'
60              will display the information about the extern step.   This  step
61              is only available when using PrologFlags=contain
62
63
64       -n, --noheader
65              No heading will be added to the output. The default action is to
66              display a header.
67
68
69       --noconvert
70              Don't convert units from their original type (e.g.  2048M  won't
71              be converted to 2G).
72
73
74       -o, --format, --fields
75              Comma  separated list of fields.  (use '--helpformat' for a list
76              of available fields).
77
78              NOTE: When using the format option for  listing  various  fields
79              you  can put a %NUMBER afterwards to specify how many characters
80              should be printed.
81
82              i.e. format=name%30 will print 30 characters of field name right
83              justified.  A -30 will print 30 characters left justified.
84
85
86       -p, --parsable
87              output will be '|' delimited with a '|' at the end
88
89
90       -P, --parsable2
91              output will be '|' delimited without a '|' at the end
92
93
94       --usage
95              Display a command usage summary.
96
97
98       -v, --verbose
99              Primarily  for  debugging  purposes, report the state of various
100              variables during processing.
101
102
103       -V, --version
104              Print version.
105
106
107
108   Job Status Fields
109       The following are the field options:
110
111              AveCPU Average (system + user) CPU time of all tasks in job.
112
113
114              AveCPUFreq
115                     Average weighted CPU frequency of all tasks  in  job,  in
116                     kHz.
117
118
119              AveDiskRead
120                     Average number of bytes read by all tasks in job.
121
122
123              AveDiskWrite
124                     Average number of bytes written by all tasks in job.
125
126
127              AvePages
128                     Average number of page faults of all tasks in job.
129
130
131              AveRSS Average resident set size of all tasks in job.
132
133
134              AveVMSize
135                     Average Virtual Memory size of all tasks in job.
136
137
138              ConsumedEnergy
139                     Total  energy  consumed  by  all tasks in job, in joules.
140                     Note: Only in case of exclusive job allocation this value
141                     reflects the jobs' real energy consumption.
142
143
144              JobID  The  number  of  the job or job step.  It is in the form:
145                     job.jobstep .
146
147
148              MaxDiskRead
149                     Maximum number of bytes read by all tasks in job.
150
151
152              MaxDiskReadNode
153                     The node on which the maxdiskread occurred.
154
155
156              MaxDiskReadTask
157                     The task ID where the maxdiskread occurred.
158
159
160              MaxDiskWrite
161                     Maximum number of bytes written by all tasks in job.
162
163
164              MaxDiskWriteNode
165                     The node on which the maxdiskwrite occurred.
166
167
168              MaxDiskWriteTask
169                     The task ID where the maxdiskwrite occurred.
170
171
172              MaxPages
173                     Maximum number of page faults of all tasks in job.
174
175
176              MaxPagesNode
177                     The node on which the maxpages occurred.
178
179
180              MaxPagesTask
181                     The task ID where the maxpages occurred.
182
183
184              MaxRSS Maximum resident set size of all tasks in job.
185
186
187              MaxRSSNode
188                     The node on which the maxrss occurred.
189
190
191              MaxRSSTask
192                     The task ID where the maxrss occurred.
193
194
195              MaxVMSize
196                     Maximum Virtual Memory size of all tasks in job.
197
198
199              MaxVMSizeNode
200                     The node on which the maxvsize occurred.
201
202
203              MaxVMSizeTask
204                     The task ID where the maxvsize occurred.
205
206
207              MinCPU Minimum (system + user) CPU time of all tasks in job.
208
209
210              MinCPUNode
211                     The node on which the mincpu occurred.
212
213
214              MinCPUTask
215                     The task ID where the mincpu occurred.
216
217
218              NTasks Total number of tasks in a job or step.
219
220
221              ReqCPUFreq
222                     Requested CPU frequency for the step, in kHz.
223
224
225              TresUsageInAve
226                     Tres average usage in by all tasks in job.  NOTE: If cor‐
227                     responding  TresUsageInMaxTask  is  -1 the metric is node
228                     centric instead of task.
229
230
231              TresUsageInMax
232                     Tres maximum usage in by all tasks in job.  NOTE: If cor‐
233                     responding  TresUsageInMaxTask  is  -1 the metric is node
234                     centric instead of task.
235
236
237              TresUsageInMaxNode
238                     Node for which each maximum TRES usage out occurred.
239
240
241              TresUsageInMaxTask
242                     Task for which each maximum TRES usage out occurred.
243
244
245              TresUsageOutAve
246                     Tres average usage out by all tasks  in  job.   NOTE:  If
247                     corresponding  TresUsageOutMaxTask  is  -1  the metric is
248                     node centric instead of task.
249
250
251              TresUsageOutMax
252                     Tres maximum usage out by all tasks  in  job.   NOTE:  If
253                     corresponding  TresUsageOutMaxTask  is  -1  the metric is
254                     node centric instead of task.
255
256
257              TresUsageOutMaxNode
258                     Node for which each maximum TRES usage out occurred.
259
260
261              TresUsageOutMaxTask
262                     Task for which each maximum TRES usage out occurred.
263
264

PERFORMANCE

266       Executing sstat sends a remote procedure call to slurmctld.  If  enough
267       calls from sstat or other Slurm client commands that send remote proce‐
268       dure calls to the slurmctld daemon come in at once, it can result in  a
269       degradation  of performance of the slurmctld daemon, possibly resulting
270       in a denial of service.
271
272       Do not run sstat or other Slurm client commands that send remote proce‐
273       dure  calls to slurmctld from loops in shell scripts or other programs.
274       Ensure that programs limit calls to sstat to the minimum necessary  for
275       the information you are trying to gather.
276
277

ENVIRONMENT VARIABLES

279       Some sstat options may be set via environment variables. These environ‐
280       ment variables, along with their corresponding options, are listed  be‐
281       low.  (Note: commandline options will always override these settings)
282
283       SLURM_CONF          The location of the Slurm configuration file.
284
285

EXAMPLES

287       Display job step information for job 11 with the specified fields:
288
289              $ sstat --format=AveCPU,AvePages,AveRSS,AveVMSize,JobID -j 11
290              25:02.000  0K         1.37M      5.93M      9.0
291
292
293       Display  job step information for job 11 with the specified fields in a
294       parsable format:
295
296              $ sstat -p --format=AveCPU,AvePages,AveRSS,AveVMSize,JobID -j 11
297              25:02.000|0K|1.37M|5.93M|9.0|
298
299

COPYING

301       Copyright (C) 2009 Lawrence Livermore National Security.   Produced  at
302       Lawrence Livermore National Laboratory (cf, DISCLAIMER).
303       Copyright (C) 2010-2013 SchedMD LLC.
304
305       This  file  is  part  of Slurm, a resource management program.  For de‐
306       tails, see <https://slurm.schedmd.com/>.
307
308       Slurm is free software; you can redistribute it and/or modify it  under
309       the  terms  of  the GNU General Public License as published by the Free
310       Software Foundation; either version 2 of the License, or (at  your  op‐
311       tion) any later version.
312
313       Slurm  is  distributed  in the hope that it will be useful, but WITHOUT
314       ANY WARRANTY; without even the implied warranty of  MERCHANTABILITY  or
315       FITNESS  FOR  A PARTICULAR PURPOSE.  See the GNU General Public License
316       for more details.
317
318

SEE ALSO

320       sacct(1)
321
322
323
324February 2021                   Slurm Commands                        sstat(1)
Impressum