1squeue(1) Slurm Commands squeue(1)
2
6 squeue - view information about jobs located in the Slurm scheduling
7 queue.
8
11 squeue [OPTIONS...]
12
15 squeue is used to view job and job step information for jobs managed by
16 Slurm.
17
20 -A, --account=<account_list>
21 Specify the accounts of the jobs to view. Accepts a comma sepa‐
22 rated list of account names. This has no effect when listing job
23 steps.
24 -a, --all
26 Display information about jobs and job steps in all partitions.
27 This causes information to be displayed about partitions that
28 are configured as hidden, partitions that are unavailable to a
29 user's group, and federated jobs that are in a "revoked" state.
30 -r, --array
32 Display one job array element per line. Without this option,
33 the display will be optimized for use with job arrays (pending
34 job array elements will be combined on one line of output with
35 the array index values printed using a regular expression).
36 --array-unique
38 Display one unique pending job array element per line. Without
39 this option, the pending job array elements will be grouped into
40 the master array job to optimize the display. This can also be
41 set with the environment variable SQUEUE_ARRAY_UNIQUE.
42 -M, --clusters=<cluster_name>
44 Clusters to issue commands to. Multiple cluster names may be
45 comma separated. A value of 'all' will query to run on all
46 clusters. This option implicitly sets the --local option.
47 --federation
49 Show jobs from the federation if a member of one.
50 -o, --format=<output_format>
52 Specify the information to be displayed, its size and position
53 (right or left justified). Also see the -O, --Format=<out‐
54 put_format> option described below (which supports less flexi‐
55 bility in formatting, but supports access to all fields). If
56 the command is executed in a federated cluster environment and
57 information about more than one cluster is to be displayed and
58 the -h, --noheader option is used, then the cluster name will be
59 displayed before the default output formats shown below.
60 The default formats with various options are:
62 default "%.18i %.9P %.8j %.8u %.2t %.10M %.6D %R"
64 -l, --long "%.18i %.9P %.8j %.8u %.8T %.10M %.9l %.6D %R"
66 -s, --steps "%.15i %.8j %.9P %.8u %.9M %N"
68 The format of each field is "%[[.]size]type[suffix]"
70 size Minimum field size. If no size is specified, whatever
72 is needed to print the information will be used.
73 . Indicates the output should be right justified and
75 size must be specified. By default output is left
76 justified.
77 suffix Arbitrary string to append to the end of the field.
79 Note that many of these type specifications are valid only for jobs
81 while others are valid only for job steps. Valid type specifications
82 include:
83 %all Print all fields available for this data type with a ver‐
85 tical bar separating each field.
86 %a Account associated with the job. (Valid for jobs only)
88 %A Number of tasks created by a job step. This reports the
90 value of the srun --ntasks option. (Valid for job steps
91 only)
92 %A Job id. This will have a unique value for each element of
94 job arrays. (Valid for jobs only)
95 %B Executing (batch) host. For an allocated session, this is
97 the host on which the session is executing (i.e. the node
98 from which the srun or the salloc command was executed).
99 For a batch job, this is the node executing the batch
100 script. In the case of a typical Linux cluster, this would
101 be the compute node zero of the allocation. In the case of
102 a Cray ALPS system, this would be the front-end host whose
103 slurmd daemon executes the job script.
104 %c Minimum number of CPUs (processors) per node requested by
106 the job. This reports the value of the srun --mincpus op‐
107 tion with a default value of zero. (Valid for jobs only)
108 %C Number of CPUs (processors) requested by the job or allo‐
110 cated to it if already running. As a job is completing
111 this number will reflect the current number of CPUs allo‐
112 cated. (Valid for jobs only)
113 %d Minimum size of temporary disk space (in MB) requested by
115 the job. (Valid for jobs only)
116 %D Number of nodes allocated to the job or the minimum number
118 of nodes required by a pending job. The actual number of
119 nodes allocated to a pending job may exceed this number if
120 the job specified a node range count (e.g. minimum and
121 maximum node counts) or the job specifies a processor
122 count instead of a node count. As a job is completing this
123 number will reflect the current number of nodes allocated.
124 (Valid for jobs only)
125 %e Time at which the job ended or is expected to end (based
127 upon its time limit). (Valid for jobs only)
128 %E Job dependencies remaining. This job will not begin execu‐
130 tion until these dependent jobs complete. In the case of a
131 job that can not run due to job dependencies never being
132 satisfied, the full original job dependency specification
133 will be reported. A value of NULL implies this job has no
134 dependencies. (Valid for jobs only)
135 %f Features required by the job. (Valid for jobs only)
137 %F Job array's job ID. This is the base job ID. For non-ar‐
139 ray jobs, this is the job ID. (Valid for jobs only)
140 %g Group name of the job. (Valid for jobs only)
142 %G Group ID of the job. (Valid for jobs only)
144 %h Can the compute resources allocated to the job be over
146 subscribed by other jobs. The resources to be over sub‐
147 scribed can be nodes, sockets, cores, or hyperthreads de‐
148 pending upon configuration. The value will be "YES" if
149 the job was submitted with the oversubscribe option or the
150 partition is configured with OverSubscribe=Force, "NO" if
151 the job requires exclusive node access, "USER" if the al‐
152 located compute nodes are dedicated to a single user,
153 "MCS" if the allocated compute nodes are dedicated to a
154 single security class (See MCSPlugin and MCSParameters
155 configuration parameters for more information), "OK" oth‐
156 erwise (typically allocated dedicated CPUs), (Valid for
157 jobs only)
158 %H Number of sockets per node requested by the job. This re‐
160 ports the value of the srun --sockets-per-node option.
161 When --sockets-per-node has not been set, "*" is dis‐
162 played. (Valid for jobs only)
163 %i Job or job step id. In the case of job arrays, the job ID
165 format will be of the form "<base_job_id>_<index>". By
166 default, the job array index field size will be limited to
167 64 bytes. Use the environment variable SLURM_BITSTR_LEN
168 to specify larger field sizes. (Valid for jobs and job
169 steps) In the case of heterogeneous job allocations, the
170 job ID format will be of the form "#+#" where the first
171 number is the "heterogeneous job leader" and the second
172 number the zero origin offset for each component of the
173 job.
174 %I Number of cores per socket requested by the job. This re‐
176 ports the value of the srun --cores-per-socket option.
177 When --cores-per-socket has not been set, "*" is dis‐
178 played. (Valid for jobs only)
179 %j Job or job step name. (Valid for jobs and job steps)
181 %J Number of threads per core requested by the job. This re‐
183 ports the value of the srun --threads-per-core option.
184 When --threads-per-core has not been set, "*" is dis‐
185 played. (Valid for jobs only)
186 %k Comment associated with the job. (Valid for jobs only)
188 %K Job array index. By default, this field size will be lim‐
190 ited to 64 bytes. Use the environment variable SLURM_BIT‐
191 STR_LEN to specify larger field sizes. (Valid for jobs
192 only)
193 %l Time limit of the job or job step in days-hours:min‐
195 utes:seconds. The value may be "NOT_SET" if not yet es‐
196 tablished or "UNLIMITED" for no limit. (Valid for jobs
197 and job steps)
198 %L Time left for the job to execute in days-hours:min‐
200 utes:seconds. This value is calculated by subtracting the
201 job's time used from its time limit. The value may be
202 "NOT_SET" if not yet established or "UNLIMITED" for no
203 limit. (Valid for jobs only)
204 %m Minimum size of memory (in MB) requested by the job.
206 (Valid for jobs only)
207 %M Time used by the job or job step in days-hours:min‐
209 utes:seconds. The days and hours are printed only as
210 needed. For job steps this field shows the elapsed time
211 since execution began and thus will be inaccurate for job
212 steps which have been suspended. Clock skew between nodes
213 in the cluster will cause the time to be inaccurate. If
214 the time is obviously wrong (e.g. negative), it displays
215 as "INVALID". (Valid for jobs and job steps)
216 %n List of node names explicitly requested by the job.
218 (Valid for jobs only)
219 %N List of nodes allocated to the job or job step. In the
221 case of a COMPLETING job, the list of nodes will comprise
222 only those nodes that have not yet been returned to ser‐
223 vice. (Valid for jobs and job steps)
224 %o The command to be executed.
226 %O Are contiguous nodes requested by the job. (Valid for
228 jobs only)
229 %p Priority of the job (converted to a floating point number
231 between 0.0 and 1.0). Also see %Q. (Valid for jobs only)
232 %P Partition of the job or job step. (Valid for jobs and job
234 steps)
235 %q Quality of service associated with the job. (Valid for
237 jobs only)
238 %Q Priority of the job (generally a very large unsigned inte‐
240 ger). Also see %p. (Valid for jobs only)
241 %r The reason a job is in its current state. See the JOB
243 REASON CODES section below for more information. (Valid
244 for jobs only)
245 %R For pending jobs: the reason a job is waiting for execu‐
247 tion is printed within parenthesis. For terminated jobs
248 with failure: an explanation as to why the job failed is
249 printed within parenthesis. For all other job states: the
250 list of allocate nodes. See the JOB REASON CODES section
251 below for more information. (Valid for jobs only)
252 %s Node selection plugin specific data for a job. Possible
254 data includes: Geometry requirement of resource allocation
255 (X,Y,Z dimensions), Connection type (TORUS, MESH, or NAV
256 == torus else mesh), Permit rotation of geometry (yes or
257 no), Node use (VIRTUAL or COPROCESSOR), etc. (Valid for
258 jobs only)
259 %S Actual or expected start time of the job or job step.
261 (Valid for jobs and job steps)
262 %t Job state in compact form. See the JOB STATE CODES sec‐
264 tion below for a list of possible states. (Valid for jobs
265 only)
266 %T Job state in extended form. See the JOB STATE CODES sec‐
268 tion below for a list of possible states. (Valid for jobs
269 only)
270 %u User name for a job or job step. (Valid for jobs and job
272 steps)
273 %U User ID for a job or job step. (Valid for jobs and job
275 steps)
276 %v Reservation for the job. (Valid for jobs only)
278 %V The job's submission time.
280 %w Workload Characterization Key (wckey). (Valid for jobs
282 only)
283 %W Licenses reserved for the job. (Valid for jobs only)
285 %x List of node names explicitly excluded by the job. (Valid
287 for jobs only)
288 %X Count of cores reserved on each node for system use (core
290 specialization). (Valid for jobs only)
291 %y Nice value (adjustment to a job's scheduling priority).
293 (Valid for jobs only)
294 %Y For pending jobs, a list of the nodes expected to be used
296 when the job is started.
297 %z Number of requested sockets, cores, and threads (S:C:T)
299 per node for the job. When (S:C:T) has not been set, "*"
300 is displayed. (Valid for jobs only)
301 %Z The job's working directory.
303 -O, --Format=<output_format>
305 Specify the information to be displayed. Also see the -o,
306 --format=<output_format> option described above (which supports
307 greater flexibility in formatting, but does not support access
308 to all fields because we ran out of letters). Requests a comma
309 separated list of job information to be displayed.
310 The format of each field is "type[:[.][size][sufix]]"
312 size Minimum field size. If no size is specified, 20 char‐
314 acters will be allocated to print the information.
315 . Indicates the output should be right justified and
317 size must be specified. By default output is left
318 justified.
319 sufix Arbitrary string to append to the end of the field.
321 Note that many of these type specifications are valid only for jobs
323 while others are valid only for job steps. Valid type specifications
324 include:
325 Account
327 Print the account associated with the job. (Valid for
328 jobs only)
329 AccrueTime
331 Print the accrue time associated with the job. (Valid
332 for jobs only)
333 admin_comment
335 Administrator comment associated with the job. (Valid
336 for jobs only)
337 AllocNodes
339 Print the nodes allocated to the job. (Valid for jobs
340 only)
341 AllocSID
343 Print the session ID used to submit the job. (Valid for
344 jobs only)
345 ArrayJobID
347 Prints the job ID of the job array. (Valid for jobs and
348 job steps)
349 ArrayTaskID
351 Prints the task ID of the job array. (Valid for jobs and
352 job steps)
353 AssocID
355 Prints the ID of the job association. (Valid for jobs
356 only)
357 BatchFlag
359 Prints whether the batch flag has been set. (Valid for
360 jobs only)
361 BatchHost
363 Executing (batch) host. For an allocated session, this is
364 the host on which the session is executing (i.e. the node
365 from which the srun or the salloc command was executed).
366 For a batch job, this is the node executing the batch
367 script. In the case of a typical Linux cluster, this
368 would be the compute node zero of the allocation. In the
369 case of a Cray ALPS system, this would be the front-end
370 host whose slurmd daemon executes the job script. (Valid
371 for jobs only)
372 BoardsPerNode
374 Prints the number of boards per node allocated to the
375 job. (Valid for jobs only)
376 BurstBuffer
378 Burst Buffer specification (Valid for jobs only)
379 BurstBufferState
381 Burst Buffer state (Valid for jobs only)
382 Cluster
384 Name of the cluster that is running the job or job step.
385 ClusterFeature
387 Cluster features required by the job. (Valid for jobs
388 only)
389 Command
391 The command to be executed. (Valid for jobs only)
392 Comment
394 Comment associated with the job. (Valid for jobs only)
395 Contiguous
397 Are contiguous nodes requested by the job. (Valid for
398 jobs only)
399 Container
401 OCI container bundle path.
402 Cores Number of cores per socket requested by the job. This
404 reports the value of the srun --cores-per-socket option.
405 When --cores-per-socket has not been set, "*" is dis‐
406 played. (Valid for jobs only)
407 CoreSpec
409 Count of cores reserved on each node for system use (core
410 specialization). (Valid for jobs only)
411 CPUFreq
413 Prints the frequency of the allocated CPUs. (Valid for
414 job steps only)
415 cpus-per-task
417 Prints the number of CPUs per tasks allocated to the job.
418 (Valid for jobs only)
419 cpus-per-tres
421 Print the memory required per trackable resources allo‐
422 cated to the job or job step.
423 Deadline
425 Prints the deadline affected to the job (Valid for jobs
426 only)
427 DelayBoot
429 Delay boot time. (Valid for jobs only)
430 Dependency
432 Job dependencies remaining. This job will not begin exe‐
433 cution until these dependent jobs complete. In the case
434 of a job that can not run due to job dependencies never
435 being satisfied, the full original job dependency speci‐
436 fication will be reported. A value of NULL implies this
437 job has no dependencies. (Valid for jobs only)
438 DerivedEC
440 Derived exit code for the job, which is the highest exit
441 code of any job step. (Valid for jobs only)
442 EligibleTime
444 Time the job is eligible for running. (Valid for jobs
445 only)
446 EndTime
448 The time of job termination, actual or expected. (Valid
449 for jobs only)
450 exit_code
452 The exit code for the job. (Valid for jobs only)
453 Feature
455 Features required by the job. (Valid for jobs only)
456 GroupID
458 Group ID of the job. (Valid for jobs only)
459 GroupName
461 Group name of the job. (Valid for jobs only)
462 HetJobID
464 Job ID of the heterogeneous job leader.
465 HetJobIDSet
467 Expression identifying all components job IDs within a
468 heterogeneous job.
469 HetJobOffset
471 Zero origin offset within a collection of heterogeneous
472 job components.
473 JobArrayID
475 Job array's job ID. This is the base job ID. For non-ar‐
476 ray jobs, this is the job ID. (Valid for jobs only)
477 JobID Job ID. This will have a unique value for each element
479 of job arrays and each component of heterogeneous jobs.
480 (Valid for jobs only)
481 LastSchedEval
483 Prints the last time the job was evaluated for schedul‐
484 ing. (Valid for jobs only)
485 Licenses
487 Licenses reserved for the job. (Valid for jobs only)
488 MaxCPUs
490 Prints the max number of CPUs allocated to the job.
491 (Valid for jobs only)
492 MaxNodes
494 Prints the max number of nodes allocated to the job.
495 (Valid for jobs only)
496 MCSLabel
498 Prints the MCS_label of the job. (Valid for jobs only)
499 mem-per-tres
501 Print the memory (in MB) required per trackable resources
502 allocated to the job or job step.
503 MinCpus
505 Minimum number of CPUs (processors) per node requested by
506 the job. This reports the value of the srun --mincpus
507 option with a default value of zero. (Valid for jobs
508 only)
509 MinMemory
511 Minimum size of memory (in MB) requested by the job.
512 (Valid for jobs only)
513 MinTime
515 Minimum time limit of the job (Valid for jobs only)
516 MinTmpDisk
518 Minimum size of temporary disk space (in MB) requested by
519 the job. (Valid for jobs only)
520 Name Job or job step name. (Valid for jobs and job steps)
522 Network
524 The network that the job is running on. (Valid for jobs
525 and job steps)
526 Nice Nice value (adjustment to a job's scheduling priority).
528 (Valid for jobs only)
529 NodeList
531 List of nodes allocated to the job or job step. In the
532 case of a COMPLETING job, the list of nodes will comprise
533 only those nodes that have not yet been returned to ser‐
534 vice. (Valid for jobs only)
535 Nodes List of nodes allocated to the job or job step. In the
537 case of a COMPLETING job, the list of nodes will comprise
538 only those nodes that have not yet been returned to ser‐
539 vice. (Valid job steps only)
540 NTPerBoard
542 The number of tasks per board allocated to the job.
543 (Valid for jobs only)
544 NTPerCore
546 The number of tasks per core allocated to the job.
547 (Valid for jobs only)
548 NTPerNode
550 The number of tasks per node allocated to the job.
551 (Valid for jobs only)
552 NTPerSocket
554 The number of tasks per socket allocated to the job.
555 (Valid for jobs only)
556 NumCPUs
558 Number of CPUs (processors) requested by the job or allo‐
559 cated to it if already running. As a job is completing,
560 this number will reflect the current number of CPUs allo‐
561 cated. (Valid for jobs and job steps)
562 NumNodes
564 Number of nodes allocated to the job or the minimum num‐
565 ber of nodes required by a pending job. The actual number
566 of nodes allocated to a pending job may exceed this num‐
567 ber if the job specified a node range count (e.g. mini‐
568 mum and maximum node counts) or the job specifies a pro‐
569 cessor count instead of a node count. As a job is com‐
570 pleting this number will reflect the current number of
571 nodes allocated. (Valid for jobs only)
572 NumTasks
574 Number of tasks requested by a job or job step. This re‐
575 ports the value of the --ntasks option. (Valid for jobs
576 and job steps)
577 Origin Cluster name where federated job originated from. (Valid
579 for federated jobs only)
580 OriginRaw
582 Cluster ID where federated job originated from. (Valid
583 for federated jobs only)
584 OverSubscribe
586 Can the compute resources allocated to the job be over
587 subscribed by other jobs. The resources to be over sub‐
588 scribed can be nodes, sockets, cores, or hyperthreads de‐
589 pending upon configuration. The value will be "YES" if
590 the job was submitted with the oversubscribe option or
591 the partition is configured with OverSubscribe=Force,
592 "NO" if the job requires exclusive node access, "USER" if
593 the allocated compute nodes are dedicated to a single
594 user, "MCS" if the allocated compute nodes are dedicated
595 to a single security class (See MCSPlugin and MCSParame‐
596 ters configuration parameters for more information), "OK"
597 otherwise (typically allocated dedicated CPUs), (Valid
598 for jobs only)
599 Partition
601 Partition of the job or job step. (Valid for jobs and
602 job steps)
603 PendingTime
605 The time (in seconds) between start time and submit time
606 of the job. If the job has not started yet, then the
607 time (in seconds) between now and the submit time of the
608 job. (Valid for jobs only)
609 PreemptTime
611 The preempt time for the job. (Valid for jobs only)
612 Prefer The preferred features of a pending job. (Valid for jobs
614 only)
615 Priority
617 Priority of the job (converted to a floating point number
618 between 0.0 and 1.0). Also see prioritylong. (Valid for
619 jobs only)
620 PriorityLong
622 Priority of the job (generally a very large unsigned in‐
623 teger). Also see priority. (Valid for jobs only)
624 Profile
626 Profile of the job. (Valid for jobs only)
627 QOS Quality of service associated with the job. (Valid for
629 jobs only)
630 Reason The reason a job is in its current state. See the JOB
632 REASON CODES section below for more information. (Valid
633 for jobs only)
634 ReasonList
636 For pending jobs: the reason a job is waiting for execu‐
637 tion is printed within parenthesis. For terminated jobs
638 with failure: an explanation as to why the job failed is
639 printed within parenthesis. For all other job states:
640 the list of allocate nodes. See the JOB REASON CODES
641 section below for more information. (Valid for jobs
642 only)
643 Reboot Indicates if the allocated nodes should be rebooted be‐
645 fore starting the job. (Valid on jobs only)
646 ReqNodes
648 List of node names explicitly requested by the job.
649 (Valid for jobs only)
650 ReqSwitch
652 The max number of requested switches by for the job.
653 (Valid for jobs only)
654 Requeue
656 Prints whether the job will be requeued on failure.
657 (Valid for jobs only)
658 Reservation
660 Reservation for the job. (Valid for jobs only)
661 ResizeTime
663 The amount of time changed for the job to run. (Valid
664 for jobs only)
665 RestartCnt
667 The number of restarts for the job. (Valid for jobs
668 only)
669 ResvPort
671 Reserved ports of the job. (Valid for job steps only)
672 SchedNodes
674 For pending jobs, a list of the nodes expected to be used
675 when the job is started. (Valid for jobs only)
676 SCT Number of requested sockets, cores, and threads (S:C:T)
678 per node for the job. When (S:C:T) has not been set, "*"
679 is displayed. (Valid for jobs only)
680 SelectJobInfo
682 Node selection plugin specific data for a job. Possible
683 data includes: Geometry requirement of resource alloca‐
684 tion (X,Y,Z dimensions), Connection type (TORUS, MESH, or
685 NAV == torus else mesh), Permit rotation of geometry (yes
686 or no), Node use (VIRTUAL or COPROCESSOR), etc. (Valid
687 for jobs only)
688 SiblingsActive
690 Cluster names of where federated sibling jobs exist.
691 (Valid for federated jobs only)
692 SiblingsActiveRaw
694 Cluster IDs of where federated sibling jobs exist.
695 (Valid for federated jobs only)
696 SiblingsViable
698 Cluster names of where federated sibling jobs are viable
699 to run. (Valid for federated jobs only)
700 SiblingsViableRaw
702 Cluster IDs of where federated sibling jobs viable to
703 run. (Valid for federated jobs only)
704 Sockets
706 Number of sockets per node requested by the job. This
707 reports the value of the srun --sockets-per-node option.
708 When --sockets-per-node has not been set, "*" is dis‐
709 played. (Valid for jobs only)
710 SPerBoard
712 Number of sockets per board allocated to the job. (Valid
713 for jobs only)
714 StartTime
716 Actual or expected start time of the job or job step.
717 (Valid for jobs and job steps)
718 State Job state in extended form. See the JOB STATE CODES sec‐
720 tion below for a list of possible states. (Valid for
721 jobs only)
722 StateCompact
724 Job state in compact form. See the JOB STATE CODES sec‐
725 tion below for a list of possible states. (Valid for
726 jobs only)
727 STDERR The directory for standard error to output to. (Valid
729 for jobs only)
730 STDIN The directory for standard in. (Valid for jobs only)
732 STDOUT The directory for standard out to output to. (Valid for
734 jobs only)
735 StepID Job or job step ID. In the case of job arrays, the job
737 ID format will be of the form "<base_job_id>_<index>".
738 (Valid forjob steps only)
739 StepName
741 Job step name. (Valid for job steps only)
742 StepState
744 The state of the job step. (Valid for job steps only)
745 SubmitTime
747 The time that the job was submitted at. (Valid for jobs
748 only)
749 system_comment
751 System comment associated with the job. (Valid for jobs
752 only)
753 Threads
755 Number of threads per core requested by the job. This
756 reports the value of the srun --threads-per-core option.
757 When --threads-per-core has not been set, "*" is dis‐
758 played. (Valid for jobs only)
759 TimeLeft
761 Time left for the job to execute in days-hours:min‐
762 utes:seconds. This value is calculated by subtracting
763 the job's time used from its time limit. The value may
764 be "NOT_SET" if not yet established or "UNLIMITED" for no
765 limit. (Valid for jobs only)
766 TimeLimit
768 Timelimit for the job or job step. (Valid for jobs and
769 job steps)
770 TimeUsed
772 Time used by the job or job step in days-hours:min‐
773 utes:seconds. The days and hours are printed only as
774 needed. For job steps this field shows the elapsed time
775 since execution began and thus will be inaccurate for job
776 steps which have been suspended. Clock skew between
777 nodes in the cluster will cause the time to be inaccu‐
778 rate. If the time is obviously wrong (e.g. negative), it
779 displays as "INVALID". (Valid for jobs and job steps)
780 tres-alloc
782 Print the trackable resources allocated to the job if
783 running. If not running, then print the trackable re‐
784 sources requested by the job.
785 tres-bind
787 Print the trackable resources task binding requested by
788 the job or job step.
789 tres-freq
791 Print the trackable resources frequencies requested by
792 the job or job step.
793 tres-per-job
795 Print the trackable resources requested by the job.
796 tres-per-node
798 Print the trackable resources per node requested by the
799 job or job step.
800 tres-per-socket
802 Print the trackable resources per socket requested by the
803 job or job step.
804 tres-per-step
806 Print the trackable resources requested by the job step.
807 tres-per-task
809 Print the trackable resources per task requested by the
810 job or job step.
811 UserID User ID for a job or job step. (Valid for jobs and job
813 steps)
814 UserName
816 User name for a job or job step. (Valid for jobs and job
817 steps)
818 Wait4Switch
820 The amount of time to wait for the desired number of
821 switches. (Valid for jobs only)
822 WCKey Workload Characterization Key (wckey). (Valid for jobs
824 only)
825 WorkDir
827 The job's working directory. (Valid for jobs only)
828 --help Print a help message describing all options squeue.
830 --hide Do not display information about jobs and job steps in all par‐
832 titions. By default, information about partitions that are con‐
833 figured as hidden or are not available to the user's group will
834 not be displayed (i.e. this is the default behavior).
835 -i, --iterate=<seconds>
837 Repeatedly gather and report the requested information at the
838 interval specified (in seconds). By default, prints a time
839 stamp with the header.
840 -j, --jobs=<job_id_list>
842 Requests a comma separated list of job IDs to display. Defaults
843 to all jobs. The --jobs=<job_id_list> option may be used in
844 conjunction with the --steps option to print step information
845 about specific jobs. Note: If a list of job IDs is provided,
846 the jobs are displayed even if they are on hidden partitions.
847 Since this option's argument is optional, for proper parsing the
848 single letter option must be followed immediately with the value
849 and not include a space between them. For example "-j1008" and
850 not "-j 1008". The job ID format is "job_id[_array_id]". Per‐
851 formance of the command can be measurably improved for systems
852 with large numbers of jobs when a single job ID is specified.
853 By default, this field size will be limited to 64 bytes. Use
854 the environment variable SLURM_BITSTR_LEN to specify larger
855 field sizes.
856 --json Dump job information as JSON. All other formatting and filtering
858 arguments will be ignored.
859 -L, --licenses=<license_list>
861 Request jobs requesting or using one or more of the named li‐
862 censes. The license list consists of a comma separated list of
863 license names.
864 --local
866 Show only jobs local to this cluster. Ignore other clusters in
867 this federation (if any). Overrides --federation.
868 -l, --long
870 Report more of the available information for the selected jobs
871 or job steps, subject to any constraints specified.
872 --me Equivalent to --user=<my username>.
874 -n, --name=<name_list>
876 Request jobs or job steps having one of the specified names.
877 The list consists of a comma separated list of job names.
878 --noconvert
880 Don't convert units from their original type (e.g. 2048M won't
881 be converted to 2G).
882 -w, --nodelist=<hostlist>
884 Report only on jobs allocated to the specified node or list of
885 nodes. This may either be the NodeName or NodeHostname as de‐
886 fined in slurm.conf(5) in the event that they differ. A
887 node_name of localhost is mapped to the current host name.
888 -h, --noheader
890 Do not print a header on the output.
891 -p, --partition=<part_list>
893 Specify the partitions of the jobs or steps to view. Accepts a
894 comma separated list of partition names.
895 -P, --priority
897 For pending jobs submitted to multiple partitions, list the job
898 once per partition. In addition, if jobs are sorted by priority,
899 consider both the partition and job priority. This option can be
900 used to produce a list of pending jobs in the same order consid‐
901 ered for scheduling by Slurm with appropriate additional options
902 (e.g. "--sort=-p,i --states=PD").
903 -q, --qos=<qos_list>
905 Specify the qos(s) of the jobs or steps to view. Accepts a comma
906 separated list of qos's.
907 -R, --reservation=<reservation_name>
909 Specify the reservation of the jobs to view.
910 --sibling
912 Show all sibling jobs on a federated cluster. Implies --federa‐
913 tion.
914 -S, --sort=<sort_list>
916 Specification of the order in which records should be reported.
917 This uses the same field specification as the <output_format>.
918 The long format option "cluster" can also be used to sort jobs
919 or job steps by cluster name (e.g. federated jobs). Multiple
920 sorts may be performed by listing multiple sort fields separated
921 by commas. The field specifications may be preceded by "+" or
922 "-" for ascending (default) and descending order respectively.
923 For example, a sort value of "P,U" will sort the records by par‐
924 tition name then by user id. The default value of sort for jobs
925 is "P,t,-p" (increasing partition name then within a given par‐
926 tition by increasing job state and then decreasing priority).
927 The default value of sort for job steps is "P,i" (increasing
928 partition name then within a given partition by increasing step
929 id).
930 --start
932 Report the expected start time and resources to be allocated for
933 pending jobs in order of increasing start time. This is equiva‐
934 lent to the following options: --format="%.18i %.9P %.8j %.8u
935 %.2t %.19S %.6D %20Y %R", --sort=S and --states=PENDING. Any
936 of these options may be explicitly changed as desired by combin‐
937 ing the --start option with other option values (e.g. to use a
938 different output format). The expected start time of pending
939 jobs is only available if the Slurm is configured to use the
940 backfill scheduling plugin.
941 -t, --states=<state_list>
943 Specify the states of jobs to view. Accepts a comma separated
944 list of state names or "all". If "all" is specified then jobs of
945 all states will be reported. If no state is specified then pend‐
946 ing, running, and completing jobs are reported. See the JOB
947 STATE CODES section below for a list of valid states. Both ex‐
948 tended and compact forms are valid. Note the <state_list> sup‐
949 plied is case insensitive ("pd" and "PD" are equivalent).
950 -s, --steps
952 Specify the job steps to view. This flag indicates that a comma
953 separated list of job steps to view follows without an equal
954 sign (see examples). The job step format is "job_id[_ar‐
955 ray_id].step_id". Defaults to all job steps. Since this option's
956 argument is optional, for proper parsing the single letter op‐
957 tion must be followed immediately with the value and not include
958 a space between them. For example "-s1008.0" and not "-s
959 1008.0".
960 --usage
962 Print a brief help message listing the squeue options.
963 -u, --user=<user_list>
965 Request jobs or job steps from a comma separated list of users.
966 The list can consist of user names or user id numbers. Perfor‐
967 mance of the command can be measurably improved for systems with
968 large numbers of jobs when a single user is specified.
969 -v, --verbose
971 Report details of squeues actions.
972 -V , --version
974 Print version information and exit.
975 --yaml Dump job information as YAML. All other formatting and filtering
977 arguments will be ignored.
978
980 These codes identify the reason that a job is waiting for execution. A
981 job may be waiting for more than one reason, in which case only one of
982 those reasons is displayed.
983 The Reasons listed below are some of the more common ones you might
985 see. For a full list of Reason codes see our Resource Limits page:
986 <https://slurm.schedmd.com/resource_limits.html>
987 AssocGrp*Limit The job's association has reached an aggregate
990 limit on some resource.
991 AssociationJobLimit The job's association has reached its maximum job
993 count.
994 AssocMax*Limit The job requests a resource that violates a per-
996 job limit on the requested association.
997 AssociationResourceLimit
999 The job's association has reached some resource
1000 limit.
1001 AssociationTimeLimit The job's association has reached its time limit.
1003 BadConstraints The job's constraints can not be satisfied.
1005 BeginTime The job's earliest start time has not yet been
1007 reached.
1008 Cleaning The job is being requeued and still cleaning up
1010 from its previous execution.
1011 Dependency This job has a dependency on another job that has
1013 not been satisfied.
1014 DependencyNeverSatisfied
1016 This job has a dependency on another job that
1017 will never be satisfied.
1018 FrontEndDown No front end node is available to execute this
1020 job.
1021 InactiveLimit The job reached the system InactiveLimit.
1023 InvalidAccount The job's account is invalid.
1025 InvalidQOS The job's QOS is invalid.
1027 JobHeldAdmin The job is held by a system administrator.
1029 JobHeldUser The job is held by the user.
1031 JobLaunchFailure The job could not be launched. This may be due
1033 to a file system problem, invalid program name,
1034 etc.
1035 Licenses The job is waiting for a license.
1037 NodeDown A node required by the job is down.
1039 NonZeroExitCode The job terminated with a non-zero exit code.
1041 PartitionDown The partition required by this job is in a DOWN
1043 state.
1044 PartitionInactive The partition required by this job is in an Inac‐
1046 tive state and not able to start jobs.
1047 PartitionNodeLimit The number of nodes required by this job is out‐
1049 side of its partition's current limits. Can also
1050 indicate that required nodes are DOWN or DRAINED.
1051 PartitionTimeLimit The job's time limit exceeds its partition's cur‐
1053 rent time limit.
1054 Priority One or more higher priority jobs exist for this
1056 partition or advanced reservation.
1057 Prolog Its PrologSlurmctld program is still running.
1059 QOSGrp*Limit The job's QOS has reached an aggregate limit on
1061 some resource.
1062 QOSJobLimit The job's QOS has reached its maximum job count.
1064 QOSMax*Limit The job requests a resource that violates a per-
1066 job limit on the requested QOS.
1067 QOSResourceLimit The job's QOS has reached some resource limit.
1069 QOSTimeLimit The job's QOS has reached its time limit.
1071 QOSUsageThreshold Required QOS threshold has been breached.
1073 ReqNodeNotAvail Some node specifically required by the job is not
1075 currently available. The node may currently be
1076 in use, reserved for another job, in an advanced
1077 reservation, DOWN, DRAINED, or not responding.
1078 Nodes which are DOWN, DRAINED, or not responding
1079 will be identified as part of the job's "reason"
1080 field as "UnavailableNodes". Such nodes will typ‐
1081 ically require the intervention of a system ad‐
1082 ministrator to make available.
1083 Reservation The job is waiting its advanced reservation to
1085 become available.
1086 Resources The job is waiting for resources to become avail‐
1088 able.
1089 SystemFailure Failure of the Slurm system, a file system, the
1091 network, etc.
1092 TimeLimit The job exhausted its time limit.
1094 WaitingForScheduling No reason has been set for this job yet. Waiting
1096 for the scheduler to determine the appropriate
1097 reason.
1098
1100 Jobs typically pass through several states in the course of their exe‐
1101 cution. The typical states are PENDING, RUNNING, SUSPENDED, COMPLET‐
1102 ING, and COMPLETED. An explanation of each state follows.
1103 BF BOOT_FAIL Job terminated due to launch failure, typically due
1106 to a hardware failure (e.g. unable to boot the node
1107 or block and the job can not be requeued).
1108 CA CANCELLED Job was explicitly cancelled by the user or system
1110 administrator. The job may or may not have been
1111 initiated.
1112 CD COMPLETED Job has terminated all processes on all nodes with
1114 an exit code of zero.
1115 CF CONFIGURING Job has been allocated resources, but are waiting
1117 for them to become ready for use (e.g. booting).
1118 CG COMPLETING Job is in the process of completing. Some processes
1120 on some nodes may still be active.
1121 DL DEADLINE Job terminated on deadline.
1123 F FAILED Job terminated with non-zero exit code or other
1125 failure condition.
1126 NF NODE_FAIL Job terminated due to failure of one or more allo‐
1128 cated nodes.
1129 OOM OUT_OF_MEMORY Job experienced out of memory error.
1131 PD PENDING Job is awaiting resource allocation.
1133 PR PREEMPTED Job terminated due to preemption.
1135 R RUNNING Job currently has an allocation.
1137 RD RESV_DEL_HOLD Job is being held after requested reservation was
1139 deleted.
1140 RF REQUEUE_FED Job is being requeued by a federation.
1142 RH REQUEUE_HOLD Held job is being requeued.
1144 RQ REQUEUED Completing job is being requeued.
1146 RS RESIZING Job is about to change size.
1148 RV REVOKED Sibling was removed from cluster due to other clus‐
1150 ter starting the job.
1151 SI SIGNALING Job is being signaled.
1153 SE SPECIAL_EXIT The job was requeued in a special state. This state
1155 can be set by users, typically in EpilogSlurmctld,
1156 if the job has terminated with a particular exit
1157 value.
1158 SO STAGE_OUT Job is staging out files.
1160 ST STOPPED Job has an allocation, but execution has been
1162 stopped with SIGSTOP signal. CPUS have been re‐
1163 tained by this job.
1164 S SUSPENDED Job has an allocation, but execution has been sus‐
1166 pended and CPUs have been released for other jobs.
1167 TO TIMEOUT Job terminated upon reaching its time limit.
1169
1171 Executing squeue sends a remote procedure call to slurmctld. If enough
1172 calls from squeue or other Slurm client commands that send remote pro‐
1173 cedure calls to the slurmctld daemon come in at once, it can result in
1174 a degradation of performance of the slurmctld daemon, possibly result‐
1175 ing in a denial of service.
1176 Do not run squeue or other Slurm client commands that send remote pro‐
1178 cedure calls to slurmctld from loops in shell scripts or other pro‐
1179 grams. Ensure that programs limit calls to squeue to the minimum neces‐
1180 sary for the information you are trying to gather.
1181
1184 Some squeue options may be set via environment variables. These envi‐
1185 ronment variables, along with their corresponding options, are listed
1186 below. (Note: Command line options will always override these set‐
1187 tings.)
1188 SLURM_BITSTR_LEN Specifies the string length to be used for holding
1191 a job array's task ID expression. The default
1192 value is 64 bytes. A value of 0 will print the
1193 full expression with any length required. Larger
1194 values may adversely impact the application perfor‐
1195 mance.
1196 SLURM_CLUSTERS Same as --clusters
1198 SLURM_CONF The location of the Slurm configuration file.
1200 SLURM_DEBUG_FLAGS Specify debug flags for squeue to use. See De‐
1202 bugFlags in the slurm.conf(5) man page for a full
1203 list of flags. The environment variable takes
1204 precedence over the setting in the slurm.conf.
1205 SLURM_TIME_FORMAT Specify the format used to report time stamps. A
1207 value of standard, the default value, generates
1208 output in the form
1209 "year-month-dateThour:minute:second". A value of
1210 relative returns only "hour:minute:second" if the
1211 current day. For other dates in the current year
1212 it prints the "hour:minute" preceded by "Tomorr"
1213 (tomorrow), "Ystday" (yesterday), the name of the
1214 day for the coming week (e.g. "Mon", "Tue", etc.),
1215 otherwise the date (e.g. "25 Apr"). For other
1216 years it returns a date month and year without a
1217 time (e.g. "6 Jun 2012"). All of the time stamps
1218 use a 24 hour format.
1219 A valid strftime() format can also be specified.
1221 For example, a value of "%a %T" will report the day
1222 of the week and a time stamp (e.g. "Mon 12:34:56").
1223 SQUEUE_ACCOUNT -A <account_list>, --account=<account_list>
1225 SQUEUE_ALL -a, --all
1227 SQUEUE_ARRAY -r, --array
1229 SQUEUE_NAMES --name=<name_list>
1231 SQUEUE_FEDERATION --federation
1233 SQUEUE_FORMAT -o <output_format>, --format=<output_format>
1235 SQUEUE_FORMAT2 -O <output_format>, --Format=<output_format>
1237 SQUEUE_LICENSES -p-l <license_list>, --license=<license_list>
1239 SQUEUE_LOCAL --local
1241 SQUEUE_PARTITION -p <part_list>, --partition=<part_list>
1243 SQUEUE_PRIORITY -P, --priority
1245 SQUEUE_QOS -p <qos_list>, --qos=<qos_list>
1247 SQUEUE_SIBLING --sibling
1249 SQUEUE_SORT -S <sort_list>, --sort=<sort_list>
1251 SQUEUE_STATES -t <state_list>, --states=<state_list>
1253 SQUEUE_USERS -u <user_list>, --users=<user_list>
1255
1257 Print the jobs scheduled in the debug partition and in the COMPLETED
1258 state in the format with six right justified digits for the job id fol‐
1259 lowed by the priority with an arbitrary fields size:
1260 $ squeue -p debug -t COMPLETED -o "%.6i %p"
1262 JOBID PRIORITY
1263 65543 99993
1264 65544 99992
1265 65545 99991
1266 Print the job steps in the debug partition sorted by user:
1269 $ squeue -s -p debug -S u
1271 STEPID NAME PARTITION USER TIME NODELIST
1272 65552.1 test1 debug alice 0:23 dev[1-4]
1273 65562.2 big_run debug bob 0:18 dev22
1274 65550.1 param1 debug candice 1:43:21 dev[6-12]
1275 Print information only about jobs 12345, 12346 and 12348:
1278 $ squeue --jobs 12345,12346,12348
1280 JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON)
1281 12345 debug job1 dave R 0:21 4 dev[9-12]
1282 12346 debug job2 dave PD 0:00 8 (Resources)
1283 12348 debug job3 ed PD 0:00 4 (Priority)
1284 Print information only about job step 65552.1:
1287 $ squeue --steps 65552.1
1289 STEPID NAME PARTITION USER TIME NODELIST
1290 65552.1 test2 debug alice 12:49 dev[1-4]
1291
1294 Copyright (C) 2002-2007 The Regents of the University of California.
1295 Produced at Lawrence Livermore National Laboratory (cf, DISCLAIMER).
1296 Copyright (C) 2008-2010 Lawrence Livermore National Security.
1297 Copyright (C) 2010-2022 SchedMD LLC.
1298 This file is part of Slurm, a resource management program. For de‐
1300 tails, see <https://slurm.schedmd.com/>.
1301 Slurm is free software; you can redistribute it and/or modify it under
1303 the terms of the GNU General Public License as published by the Free
1304 Software Foundation; either version 2 of the License, or (at your op‐
1305 tion) any later version.
1306 Slurm is distributed in the hope that it will be useful, but WITHOUT
1308 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
1309 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
1310 for more details.
1311
1313 scancel(1), scontrol(1), sinfo(1), srun(1), slurm_load_ctl_conf [22m(3),
1314 slurm_load_jobs (3), slurm_load_node (3), slurm_load_partitions (3)
1315October 2022 Slurm Commands squeue(1)