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_list>, --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
27 Display information about jobs and job steps in all partitions.
28 This causes information to be displayed about partitions that
29 are configured as hidden and partitions that are unavailable to
30 user's group.
31 -r, --array
34 Display one job array element per line. Without this option,
35 the display will be optimized for use with job arrays (pending
36 job array elements will be combined on one line of output with
37 the array index values printed using a regular expression).
38 --array-unique
41 Display one unique pending job array element per line. Without
42 this option, the pending job array elements will be grouped into
43 the master array job to optimize the display. This can also be
44 set with the environment variable SQUEUE_ARRAY_UNIQUE.
45 --federation
48 Show jobs from the federation if a member of one.
49 -h, --noheader
52 Do not print a header on the output.
53 --help Print a help message describing all options squeue.
56 --hide Do not display information about jobs and job steps in all par‐
59 titions. By default, information about partitions that are con‐
60 figured as hidden or are not available to the user's group will
61 not be displayed (i.e. this is the default behavior).
62 -i <seconds>, --iterate=<seconds>
65 Repeatedly gather and report the requested information at the
66 interval specified (in seconds). By default, prints a time
67 stamp with the header.
68 -j <job_id_list>, --jobs=<job_id_list>
71 Requests a comma separated list of job IDs to display. Defaults
72 to all jobs. The --jobs=<job_id_list> option may be used in
73 conjunction with the --steps option to print step information
74 about specific jobs. Note: If a list of job IDs is provided,
75 the jobs are displayed even if they are on hidden partitions.
76 Since this option's argument is optional, for proper parsing the
77 single letter option must be followed immediately with the value
78 and not include a space between them. For example "-j1008" and
79 not "-j 1008". The job ID format is "job_id[_array_id]". Per‐
80 formance of the command can be measurably improved for systems
81 with large numbers of jobs when a single job ID is specified.
82 By default, this field size will be limited to 64 bytes. Use
83 the environment variable SLURM_BITSTR_LEN to specify larger
84 field sizes.
85 --local
89 Show only jobs local to this cluster. Ignore other clusters in
90 this federation (if any). Overrides --federation.
91 -l, --long
94 Report more of the available information for the selected jobs
95 or job steps, subject to any constraints specified.
96 -L, --licenses=<license_list>
99 Request jobs requesting or using one or more of the named
100 licenses. The license list consists of a comma separated list
101 of license names.
102 --me Equivalent to --user=<my username>.
105 -M, --clusters=<cluster_name>
108 Clusters to issue commands to. Multiple cluster names may be
109 comma separated. A value of 'all' will query to run on all
110 clusters. This option implicitly sets the --local option.
111 -n, --name=<name_list>
114 Request jobs or job steps having one of the specified names.
115 The list consists of a comma separated list of job names.
116 --noconvert
119 Don't convert units from their original type (e.g. 2048M won't
120 be converted to 2G).
121 -o <output_format>, --format=<output_format>
124 Specify the information to be displayed, its size and position
125 (right or left justified). Also see the -O <output_format>,
126 --Format=<output_format> option described below (which supports
127 less flexibility in formatting, but supports access to all
128 fields). If the command is executed in a federated cluster
129 environment and information about more than one cluster is to be
130 displayed and the -h, --noheader option is used, then the clus‐
131 ter name will be displayed before the default output formats
132 shown below.
133 The default formats with various options are:
135 default "%.18i %.9P %.8j %.8u %.2t %.10M %.6D %R"
137 -l, --long "%.18i %.9P %.8j %.8u %.8T %.10M %.9l %.6D %R"
139 -s, --steps "%.15i %.8j %.9P %.8u %.9M %N"
141 The format of each field is "%[[.]size]type[suffix]"
144 size Minimum field size. If no size is specified, whatever
146 is needed to print the information will be used.
147 . Indicates the output should be right justified and
149 size must be specified. By default output is left
150 justified.
151 suffix Arbitrary string to append to the end of the field.
153 Note that many of these type specifications are valid only for
156 jobs while others are valid only for job steps. Valid type
157 specifications include:
158 %all Print all fields available for this data type with a ver‐
161 tical bar separating each field.
162 %a Account associated with the job. (Valid for jobs only)
164 %A Number of tasks created by a job step. This reports the
166 value of the srun --ntasks option. (Valid for job steps
167 only)
168 %A Job id. This will have a unique value for each element of
170 job arrays. (Valid for jobs only)
171 %B Executing (batch) host. For an allocated session, this is
173 the host on which the session is executing (i.e. the node
174 from which the srun or the salloc command was executed).
175 For a batch job, this is the node executing the batch
176 script. In the case of a typical Linux cluster, this would
177 be the compute node zero of the allocation. In the case of
178 a Cray ALPS system, this would be the front-end host whose
179 slurmd daemon executes the job script.
180 %c Minimum number of CPUs (processors) per node requested by
182 the job. This reports the value of the srun --mincpus
183 option with a default value of zero. (Valid for jobs
184 only)
185 %C Number of CPUs (processors) requested by the job or allo‐
187 cated to it if already running. As a job is completing
188 this number will reflect the current number of CPUs allo‐
189 cated. (Valid for jobs only)
190 %d Minimum size of temporary disk space (in MB) requested by
192 the job. (Valid for jobs only)
193 %D Number of nodes allocated to the job or the minimum number
195 of nodes required by a pending job. The actual number of
196 nodes allocated to a pending job may exceed this number if
197 the job specified a node range count (e.g. minimum and
198 maximum node counts) or the job specifies a processor
199 count instead of a node count. As a job is completing this
200 number will reflect the current number of nodes allocated.
201 (Valid for jobs only)
202 %e Time at which the job ended or is expected to end (based
204 upon its time limit). (Valid for jobs only)
205 %E Job dependencies remaining. This job will not begin execu‐
207 tion until these dependent jobs complete. In the case of a
208 job that can not run due to job dependencies never being
209 satisfied, the full original job dependency specification
210 will be reported. A value of NULL implies this job has no
211 dependencies. (Valid for jobs only)
212 %f Features required by the job. (Valid for jobs only)
214 %F Job array's job ID. This is the base job ID. For
216 non-array jobs, this is the job ID. (Valid for jobs only)
217 %g Group name of the job. (Valid for jobs only)
219 %G Group ID of the job. (Valid for jobs only)
221 %h Can the compute resources allocated to the job be over
223 subscribed by other jobs. The resources to be over sub‐
224 scribed can be nodes, sockets, cores, or hyperthreads
225 depending upon configuration. The value will be "YES" if
226 the job was submitted with the oversubscribe option or the
227 partition is configured with OverSubscribe=Force, "NO" if
228 the job requires exclusive node access, "USER" if the
229 allocated compute nodes are dedicated to a single user,
230 "MCS" if the allocated compute nodes are dedicated to a
231 single security class (See MCSPlugin and MCSParameters
232 configuration parameters for more information), "OK" oth‐
233 erwise (typically allocated dedicated CPUs), (Valid for
234 jobs only)
235 %H Number of sockets per node requested by the job. This
237 reports the value of the srun --sockets-per-node option.
238 When --sockets-per-node has not been set, "*" is dis‐
239 played. (Valid for jobs only)
240 %i Job or job step id. In the case of job arrays, the job ID
242 format will be of the form "<base_job_id>_<index>". By
243 default, the job array index field size will be limited to
244 64 bytes. Use the environment variable SLURM_BITSTR_LEN
245 to specify larger field sizes. (Valid for jobs and job
246 steps) In the case of heterogeneous job allocations, the
247 job ID format will be of the form "#+#" where the first
248 number is the "heterogeneous job leader" and the second
249 number the zero origin offset for each component of the
250 job.
251 %I Number of cores per socket requested by the job. This
253 reports the value of the srun --cores-per-socket option.
254 When --cores-per-socket has not been set, "*" is dis‐
255 played. (Valid for jobs only)
256 %j Job or job step name. (Valid for jobs and job steps)
258 %J Number of threads per core requested by the job. This
260 reports the value of the srun --threads-per-core option.
261 When --threads-per-core has not been set, "*" is dis‐
262 played. (Valid for jobs only)
263 %k Comment associated with the job. (Valid for jobs only)
265 %K Job array index. By default, this field size will be lim‐
267 ited to 64 bytes. Use the environment variable SLURM_BIT‐
268 STR_LEN to specify larger field sizes. (Valid for jobs
269 only)
270 %l Time limit of the job or job step in days-hours:min‐
272 utes:seconds. The value may be "NOT_SET" if not yet
273 established or "UNLIMITED" for no limit. (Valid for jobs
274 and job steps)
275 %L Time left for the job to execute in days-hours:min‐
277 utes:seconds. This value is calculated by subtracting the
278 job's time used from its time limit. The value may be
279 "NOT_SET" if not yet established or "UNLIMITED" for no
280 limit. (Valid for jobs only)
281 %m Minimum size of memory (in MB) requested by the job.
283 (Valid for jobs only)
284 %M Time used by the job or job step in days-hours:min‐
286 utes:seconds. The days and hours are printed only as
287 needed. For job steps this field shows the elapsed time
288 since execution began and thus will be inaccurate for job
289 steps which have been suspended. Clock skew between nodes
290 in the cluster will cause the time to be inaccurate. If
291 the time is obviously wrong (e.g. negative), it displays
292 as "INVALID". (Valid for jobs and job steps)
293 %n List of node names explicitly requested by the job.
295 (Valid for jobs only)
296 %N List of nodes allocated to the job or job step. In the
298 case of a COMPLETING job, the list of nodes will comprise
299 only those nodes that have not yet been returned to ser‐
300 vice. (Valid for jobs and job steps)
301 %o The command to be executed.
303 %O Are contiguous nodes requested by the job. (Valid for
305 jobs only)
306 %p Priority of the job (converted to a floating point number
308 between 0.0 and 1.0). Also see %Q. (Valid for jobs only)
309 %P Partition of the job or job step. (Valid for jobs and job
311 steps)
312 %q Quality of service associated with the job. (Valid for
314 jobs only)
315 %Q Priority of the job (generally a very large unsigned inte‐
317 ger). Also see %p. (Valid for jobs only)
318 %r The reason a job is in its current state. See the JOB
320 REASON CODES section below for more information. (Valid
321 for jobs only)
322 %R For pending jobs: the reason a job is waiting for execu‐
324 tion is printed within parenthesis. For terminated jobs
325 with failure: an explanation as to why the job failed is
326 printed within parenthesis. For all other job states: the
327 list of allocate nodes. See the JOB REASON CODES section
328 below for more information. (Valid for jobs only)
329 %s Node selection plugin specific data for a job. Possible
331 data includes: Geometry requirement of resource allocation
332 (X,Y,Z dimensions), Connection type (TORUS, MESH, or NAV
333 == torus else mesh), Permit rotation of geometry (yes or
334 no), Node use (VIRTUAL or COPROCESSOR), etc. (Valid for
335 jobs only)
336 %S Actual or expected start time of the job or job step.
338 (Valid for jobs and job steps)
339 %t Job state in compact form. See the JOB STATE CODES sec‐
341 tion below for a list of possible states. (Valid for jobs
342 only)
343 %T Job state in extended form. See the JOB STATE CODES sec‐
345 tion below for a list of possible states. (Valid for jobs
346 only)
347 %u User name for a job or job step. (Valid for jobs and job
349 steps)
350 %U User ID for a job or job step. (Valid for jobs and job
352 steps)
353 %v Reservation for the job. (Valid for jobs only)
355 %V The job's submission time.
357 %w Workload Characterization Key (wckey). (Valid for jobs
359 only)
360 %W Licenses reserved for the job. (Valid for jobs only)
362 %x List of node names explicitly excluded by the job. (Valid
364 for jobs only)
365 %X Count of cores reserved on each node for system use (core
367 specialization). (Valid for jobs only)
368 %y Nice value (adjustment to a job's scheduling priority).
370 (Valid for jobs only)
371 %Y For pending jobs, a list of the nodes expected to be used
373 when the job is started.
374 %z Number of requested sockets, cores, and threads (S:C:T)
376 per node for the job. When (S:C:T) has not been set, "*"
377 is displayed. (Valid for jobs only)
378 %Z The job's working directory.
380 -O <output_format>, --Format=<output_format>
384 Specify the information to be displayed. Also see the -o <out‐
385 put_format>, --format=<output_format> option described below
386 (which supports greater flexibility in formatting, but does not
387 support access to all fields because we ran out of letters).
388 Requests a comma separated list of job information to be dis‐
389 played.
390 The format of each field is "type[:[.][size][sufix]]"
393 size Minimum field size. If no size is specified, 20 char‐
395 acters will be allocated to print the information.
396 . Indicates the output should be right justified and
398 size must be specified. By default output is left
399 justified.
400 sufix Arbitrary string to append to the end of the field.
402 Note that many of these type specifications are valid only for
405 jobs while others are valid only for job steps. Valid type
406 specifications include:
407 Account
410 Print the account associated with the job. (Valid for
411 jobs only)
412 AccrueTime
414 Print the accrue time associated with the job. (Valid
415 for jobs only)
416 admin_comment
418 Administrator comment associated with the job. (Valid
419 for jobs only)
420 AllocNodes
422 Print the nodes allocated to the job. (Valid for jobs
423 only)
424 AllocSID
426 Print the session ID used to submit the job. (Valid for
427 jobs only)
428 ArrayJobID
430 Prints the job ID of the job array. (Valid for jobs and
431 job steps)
432 ArrayTaskID
434 Prints the task ID of the job array. (Valid for jobs and
435 job steps)
436 AssocID
438 Prints the ID of the job association. (Valid for jobs
439 only)
440 BatchFlag
442 Prints whether the batch flag has been set. (Valid for
443 jobs only)
444 BatchHost
446 Executing (batch) host. For an allocated session, this is
447 the host on which the session is executing (i.e. the node
448 from which the srun or the salloc command was executed).
449 For a batch job, this is the node executing the batch
450 script. In the case of a typical Linux cluster, this
451 would be the compute node zero of the allocation. In the
452 case of a Cray ALPS system, this would be the front-end
453 host whose slurmd daemon executes the job script. (Valid
454 for jobs only)
455 BoardsPerNode
457 Prints the number of boards per node allocated to the
458 job. (Valid for jobs only)
459 BurstBuffer
461 Burst Buffer specification (Valid for jobs only)
462 BurstBufferState
464 Burst Buffer state (Valid for jobs only)
465 Cluster
467 Name of the cluster that is running the job or job step.
468 ClusterFeature
470 Cluster features required by the job. (Valid for jobs
471 only)
472 Command
474 The command to be executed. (Valid for jobs only)
475 Comment
477 Comment associated with the job. (Valid for jobs only)
478 Contiguous
480 Are contiguous nodes requested by the job. (Valid for
481 jobs only)
482 Cores
484 Number of cores per socket requested by the job. This
485 reports the value of the srun --cores-per-socket option.
486 When --cores-per-socket has not been set, "*" is dis‐
487 played. (Valid for jobs only)
488 CoreSpec
490 Count of cores reserved on each node for system use (core
491 specialization). (Valid for jobs only)
492 CPUFreq
494 Prints the frequency of the allocated CPUs. (Valid for
495 job steps only)
496 cpus-per-task
498 Prints the number of CPUs per tasks allocated to the job.
499 (Valid for jobs only)
500 cpus-per-tres
502 Print the memory required per trackable resources allo‐
503 cated to the job or job step.
504 Deadline
506 Prints the deadline affected to the job (Valid for jobs
507 only)
508 DelayBoot
510 Delay boot time. (Valid for jobs only)
511 Dependency
513 Job dependencies remaining. This job will not begin exe‐
514 cution until these dependent jobs complete. In the case
515 of a job that can not run due to job dependencies never
516 being satisfied, the full original job dependency speci‐
517 fication will be reported. A value of NULL implies this
518 job has no dependencies. (Valid for jobs only)
519 DerivedEC
521 Derived exit code for the job, which is the highest exit
522 code of any job step. (Valid for jobs only)
523 EligibleTime
525 Time the job is eligible for running. (Valid for jobs
526 only)
527 EndTime
529 The time of job termination, actual or expected. (Valid
530 for jobs only)
531 exit_code
533 The exit code for the job. (Valid for jobs only)
534 Feature
536 Features required by the job. (Valid for jobs only)
537 GroupID
539 Group ID of the job. (Valid for jobs only)
540 GroupName
542 Group name of the job. (Valid for jobs only)
543 HetJobID
545 Job ID of the heterogeneous job leader.
546 HetJobIDSet
548 Expression identifying all components job IDs within a
549 heterogeneous job.
550 HetJobOffset
552 Zero origin offset within a collection of heterogeneous
553 job components.
554 JobArrayID
556 Job array's job ID. This is the base job ID. For
557 non-array jobs, this is the job ID. (Valid for jobs
558 only)
559 JobID
561 Job ID. This will have a unique value for each element
562 of job arrays and each component of heterogeneous jobs.
563 (Valid for jobs only)
564 LastSchedEval
566 Prints the last time the job was evaluated for schedul‐
567 ing. (Valid for jobs only)
568 Licenses
570 Licenses reserved for the job. (Valid for jobs only)
571 MaxCPUs
573 Prints the max number of CPUs allocated to the job.
574 (Valid for jobs only)
575 MaxNodes
577 Prints the max number of nodes allocated to the job.
578 (Valid for jobs only)
579 MCSLabel
581 Prints the MCS_label of the job. (Valid for jobs only)
582 mem-per-tres
584 Print the memory (in MB) required per trackable resources
585 allocated to the job or job step.
586 MinCpus
588 Minimum number of CPUs (processors) per node requested by
589 the job. This reports the value of the srun --mincpus
590 option with a default value of zero. (Valid for jobs
591 only)
592 MinMemory
594 Minimum size of memory (in MB) requested by the job.
595 (Valid for jobs only)
596 MinTime
598 Minimum time limit of the job (Valid for jobs only)
599 MinTmpDisk
601 Minimum size of temporary disk space (in MB) requested by
602 the job. (Valid for jobs only)
603 Name
605 Job or job step name. (Valid for jobs and job steps)
606 Network
608 The network that the job is running on. (Valid for jobs
609 and job steps)
610 Nice
612 Nice value (adjustment to a job's scheduling priority).
613 (Valid for jobs only)
614 NodeList
616 List of nodes allocated to the job or job step. In the
617 case of a COMPLETING job, the list of nodes will comprise
618 only those nodes that have not yet been returned to ser‐
619 vice. (Valid for jobs only)
620 Nodes
622 List of nodes allocated to the job or job step. In the
623 case of a COMPLETING job, the list of nodes will comprise
624 only those nodes that have not yet been returned to ser‐
625 vice. (Valid job steps only)
626 NTPerBoard
628 The number of tasks per board allocated to the job.
629 (Valid for jobs only)
630 NTPerCore
632 The number of tasks per core allocated to the job.
633 (Valid for jobs only)
634 NTPerNode
636 The number of tasks per node allocated to the job.
637 (Valid for jobs only)
638 NTPerSocket
640 The number of tasks per socket allocated to the job.
641 (Valid for jobs only)
642 NumCPUs
644 Number of CPUs (processors) requested by the job or allo‐
645 cated to it if already running. As a job is completing,
646 this number will reflect the current number of CPUs allo‐
647 cated. (Valid for jobs and job steps)
648 NumNodes
650 Number of nodes allocated to the job or the minimum num‐
651 ber of nodes required by a pending job. The actual number
652 of nodes allocated to a pending job may exceed this num‐
653 ber if the job specified a node range count (e.g. mini‐
654 mum and maximum node counts) or the job specifies a pro‐
655 cessor count instead of a node count. As a job is com‐
656 pleting this number will reflect the current number of
657 nodes allocated. (Valid for jobs only)
658 NumTasks
660 Number of tasks requested by a job or job step. This
661 reports the value of the --ntasks option. (Valid for
662 jobs and job steps)
663 Origin
665 Cluster name where federated job originated from. (Valid
666 for federated jobs only)
667 OriginRaw
669 Cluster ID where federated job originated from. (Valid
670 for federated jobs only)
671 OverSubscribe
673 Can the compute resources allocated to the job be over
674 subscribed by other jobs. The resources to be over sub‐
675 scribed can be nodes, sockets, cores, or hyperthreads
676 depending upon configuration. The value will be "YES" if
677 the job was submitted with the oversubscribe option or
678 the partition is configured with OverSubscribe=Force,
679 "NO" if the job requires exclusive node access, "USER" if
680 the allocated compute nodes are dedicated to a single
681 user, "MCS" if the allocated compute nodes are dedicated
682 to a single security class (See MCSPlugin and MCSParame‐
683 ters configuration parameters for more information), "OK"
684 otherwise (typically allocated dedicated CPUs), (Valid
685 for jobs only)
686 Partition
688 Partition of the job or job step. (Valid for jobs and
689 job steps)
690 PreemptTime
692 The preempt time for the job. (Valid for jobs only)
693 PendingTime
695 The time (in seconds) between start time and submit time
696 of the job. If the job has not started yet, then the
697 time (in seconds) between now and the submit time of the
698 job. (Valid for jobs only)
699 Priority
701 Priority of the job (converted to a floating point number
702 between 0.0 and 1.0). Also see prioritylong. (Valid for
703 jobs only)
704 PriorityLong
706 Priority of the job (generally a very large unsigned
707 integer). Also see priority. (Valid for jobs only)
708 Profile
710 Profile of the job. (Valid for jobs only)
711 QOS
713 Quality of service associated with the job. (Valid for
714 jobs only)
715 Reason
717 The reason a job is in its current state. See the JOB
718 REASON CODES section below for more information. (Valid
719 for jobs only)
720 ReasonList
722 For pending jobs: the reason a job is waiting for execu‐
723 tion is printed within parenthesis. For terminated jobs
724 with failure: an explanation as to why the job failed is
725 printed within parenthesis. For all other job states:
726 the list of allocate nodes. See the JOB REASON CODES
727 section below for more information. (Valid for jobs
728 only)
729 Reboot
731 Indicates if the allocated nodes should be rebooted
732 before starting the job. (Valid on jobs only)
733 ReqNodes
735 List of node names explicitly requested by the job.
736 (Valid for jobs only)
737 ReqSwitch
739 The max number of requested switches by for the job.
740 (Valid for jobs only)
741 Requeue
743 Prints whether the job will be requeued on failure.
744 (Valid for jobs only)
745 Reservation
747 Reservation for the job. (Valid for jobs only)
748 ResizeTime
750 The amount of time changed for the job to run. (Valid
751 for jobs only)
752 RestartCnt
754 The number of restarts for the job. (Valid for jobs
755 only)
756 ResvPort
758 Reserved ports of the job. (Valid for job steps only)
759 SchedNodes
761 For pending jobs, a list of the nodes expected to be used
762 when the job is started. (Valid for jobs only)
763 SCT
765 Number of requested sockets, cores, and threads (S:C:T)
766 per node for the job. When (S:C:T) has not been set, "*"
767 is displayed. (Valid for jobs only)
768 SelectJobInfo
770 Node selection plugin specific data for a job. Possible
771 data includes: Geometry requirement of resource alloca‐
772 tion (X,Y,Z dimensions), Connection type (TORUS, MESH, or
773 NAV == torus else mesh), Permit rotation of geometry (yes
774 or no), Node use (VIRTUAL or COPROCESSOR), etc. (Valid
775 for jobs only)
776 SiblingsActive
778 Cluster names of where federated sibling jobs exist.
779 (Valid for federated jobs only)
780 SiblingsActiveRaw
782 Cluster IDs of where federated sibling jobs exist.
783 (Valid for federated jobs only)
784 SiblingsViable
786 Cluster names of where federated sibling jobs are viable
787 to run. (Valid for federated jobs only)
788 SiblingsViableRaw
790 Cluster IDs of where federated sibling jobs viable to
791 run. (Valid for federated jobs only)
792 Sockets
794 Number of sockets per node requested by the job. This
795 reports the value of the srun --sockets-per-node option.
796 When --sockets-per-node has not been set, "*" is dis‐
797 played. (Valid for jobs only)
798 SPerBoard
800 Number of sockets per board allocated to the job. (Valid
801 for jobs only)
802 StartTime
804 Actual or expected start time of the job or job step.
805 (Valid for jobs and job steps)
806 State
808 Job state in extended form. See the JOB STATE CODES sec‐
809 tion below for a list of possible states. (Valid for
810 jobs only)
811 StateCompact
813 Job state in compact form. See the JOB STATE CODES sec‐
814 tion below for a list of possible states. (Valid for
815 jobs only)
816 STDERR
818 The directory for standard error to output to. (Valid
819 for jobs only)
820 STDIN
822 The directory for standard in. (Valid for jobs only)
823 STDOUT
825 The directory for standard out to output to. (Valid for
826 jobs only)
827 StepID
829 Job or job step ID. In the case of job arrays, the job
830 ID format will be of the form "<base_job_id>_<index>".
831 (Valid forjob steps only)
832 StepName
834 Job step name. (Valid for job steps only)
835 StepState
837 The state of the job step. (Valid for job steps only)
838 SubmitTime
840 The time that the job was submitted at. (Valid for jobs
841 only)
842 system_comment
844 System comment associated with the job. (Valid for jobs
845 only)
846 Threads
848 Number of threads per core requested by the job. This
849 reports the value of the srun --threads-per-core option.
850 When --threads-per-core has not been set, "*" is dis‐
851 played. (Valid for jobs only)
852 TimeLeft
854 Time left for the job to execute in days-hours:min‐
855 utes:seconds. This value is calculated by subtracting
856 the job's time used from its time limit. The value may
857 be "NOT_SET" if not yet established or "UNLIMITED" for no
858 limit. (Valid for jobs only)
859 TimeLimit
861 Timelimit for the job or job step. (Valid for jobs and
862 job steps)
863 TimeUsed
865 Time used by the job or job step in days-hours:min‐
866 utes:seconds. The days and hours are printed only as
867 needed. For job steps this field shows the elapsed time
868 since execution began and thus will be inaccurate for job
869 steps which have been suspended. Clock skew between
870 nodes in the cluster will cause the time to be inaccu‐
871 rate. If the time is obviously wrong (e.g. negative), it
872 displays as "INVALID". (Valid for jobs and job steps)
873 tres-alloc
875 Print the trackable resources allocated to the job if
876 running. If not running, then print the trackable
877 resources requested by the job.
878 tres-bind
880 Print the trackable resources task binding requested by
881 the job or job step.
882 tres-freq
884 Print the trackable resources frequencies requested by
885 the job or job step.
886 tres-per-job
888 Print the trackable resources requested by the job.
889 tres-per-node
891 Print the trackable resources per node requested by the
892 job or job step.
893 tres-per-socket
895 Print the trackable resources per socket requested by the
896 job or job step.
897 tres-per-step
899 Print the trackable resources requested by the job step.
900 tres-per-task
902 Print the trackable resources per task requested by the
903 job or job step.
904 UserID
906 User ID for a job or job step. (Valid for jobs and job
907 steps)
908 UserName
910 User name for a job or job step. (Valid for jobs and job
911 steps)
912 Wait4Switch
914 The amount of time to wait for the desired number of
915 switches. (Valid for jobs only)
916 WCKey
918 Workload Characterization Key (wckey). (Valid for jobs
919 only)
920 WorkDir
922 The job's working directory. (Valid for jobs only)
923 -p <part_list>, --partition=<part_list>
926 Specify the partitions of the jobs or steps to view. Accepts a
927 comma separated list of partition names.
928 -P, --priority
931 For pending jobs submitted to multiple partitions, list the job
932 once per partition. In addition, if jobs are sorted by priority,
933 consider both the partition and job priority. This option can be
934 used to produce a list of pending jobs in the same order consid‐
935 ered for scheduling by Slurm with appropriate additional options
936 (e.g. "--sort=-p,i --states=PD").
937 -q <qos_list>, --qos=<qos_list>
940 Specify the qos(s) of the jobs or steps to view. Accepts a comma
941 separated list of qos's.
942 -R, --reservation=<reservation_name>
945 Specify the reservation of the jobs to view.
946 -s, --steps
949 Specify the job steps to view. This flag indicates that a comma
950 separated list of job steps to view follows without an equal
951 sign (see examples). The job step format is
952 "job_id[_array_id].step_id". Defaults to all job steps. Since
953 this option's argument is optional, for proper parsing the sin‐
954 gle letter option must be followed immediately with the value
955 and not include a space between them. For example "-s1008.0" and
956 not "-s 1008.0".
957 --sibling
960 Show all sibling jobs on a federated cluster. Implies --federa‐
961 tion.
962 -S <sort_list>, --sort=<sort_list>
965 Specification of the order in which records should be reported.
966 This uses the same field specification as the <output_format>.
967 The long format option "cluster" can also be used to sort jobs
968 or job steps by cluster name (e.g. federated jobs). Multiple
969 sorts may be performed by listing multiple sort fields separated
970 by commas. The field specifications may be preceded by "+" or
971 "-" for ascending (default) and descending order respectively.
972 For example, a sort value of "P,U" will sort the records by par‐
973 tition name then by user id. The default value of sort for jobs
974 is "P,t,-p" (increasing partition name then within a given par‐
975 tition by increasing job state and then decreasing priority).
976 The default value of sort for job steps is "P,i" (increasing
977 partition name then within a given partition by increasing step
978 id).
979 --start
982 Report the expected start time and resources to be allocated for
983 pending jobs in order of increasing start time. This is equiva‐
984 lent to the following options: --format="%.18i %.9P %.8j %.8u
985 %.2t %.19S %.6D %20Y %R", --sort=S and --states=PENDING. Any
986 of these options may be explicitly changed as desired by combin‐
987 ing the --start option with other option values (e.g. to use a
988 different output format). The expected start time of pending
989 jobs is only available if the Slurm is configured to use the
990 backfill scheduling plugin.
991 -t <state_list>, --states=<state_list>
994 Specify the states of jobs to view. Accepts a comma separated
995 list of state names or "all". If "all" is specified then jobs of
996 all states will be reported. If no state is specified then pend‐
997 ing, running, and completing jobs are reported. See the JOB
998 STATE CODES section below for a list of valid states. Both
999 extended and compact forms are valid. Note the <state_list>
1000 supplied is case insensitive ("pd" and "PD" are equivalent).
1001 -u <user_list>, --user=<user_list>
1004 Request jobs or job steps from a comma separated list of users.
1005 The list can consist of user names or user id numbers. Perfor‐
1006 mance of the command can be measurably improved for systems with
1007 large numbers of jobs when a single user is specified.
1008 --usage
1011 Print a brief help message listing the squeue options.
1012 -v, --verbose
1015 Report details of squeues actions.
1016 -V , --version
1019 Print version information and exit.
1020 -w <hostlist>, --nodelist=<hostlist>
1023 Report only on jobs allocated to the specified node or list of
1024 nodes. This may either be the NodeName or NodeHostname as
1025 defined in slurm.conf(5) in the event that they differ. A
1026 node_name of localhost is mapped to the current host name.
1027
1030 These codes identify the reason that a job is waiting for execution. A
1031 job may be waiting for more than one reason, in which case only one of
1032 those reasons is displayed.
1033 AssociationJobLimit The job's association has reached its maximum job
1035 count.
1036 AssociationResourceLimit
1038 The job's association has reached some resource
1039 limit.
1040 AssociationTimeLimit The job's association has reached its time limit.
1042 BadConstraints The job's constraints can not be satisfied.
1044 BeginTime The job's earliest start time has not yet been
1046 reached.
1047 Cleaning The job is being requeued and still cleaning up
1049 from its previous execution.
1050 Dependency This job is waiting for a dependent job to com‐
1052 plete.
1053 FrontEndDown No front end node is available to execute this
1055 job.
1056 InactiveLimit The job reached the system InactiveLimit.
1058 InvalidAccount The job's account is invalid.
1060 InvalidQOS The job's QOS is invalid.
1062 JobHeldAdmin The job is held by a system administrator.
1064 JobHeldUser The job is held by the user.
1066 JobLaunchFailure The job could not be launched. This may be due
1068 to a file system problem, invalid program name,
1069 etc.
1070 Licenses The job is waiting for a license.
1072 NodeDown A node required by the job is down.
1074 NonZeroExitCode The job terminated with a non-zero exit code.
1076 PartitionDown The partition required by this job is in a DOWN
1078 state.
1079 PartitionInactive The partition required by this job is in an Inac‐
1081 tive state and not able to start jobs.
1082 PartitionNodeLimit The number of nodes required by this job is out‐
1084 side of its partition's current limits. Can also
1085 indicate that required nodes are DOWN or DRAINED.
1086 PartitionTimeLimit The job's time limit exceeds its partition's cur‐
1088 rent time limit.
1089 Priority One or more higher priority jobs exist for this
1091 partition or advanced reservation.
1092 Prolog Its PrologSlurmctld program is still running.
1094 QOSJobLimit The job's QOS has reached its maximum job count.
1096 QOSResourceLimit The job's QOS has reached some resource limit.
1098 QOSTimeLimit The job's QOS has reached its time limit.
1100 ReqNodeNotAvail Some node specifically required by the job is not
1102 currently available. The node may currently be
1103 in use, reserved for another job, in an advanced
1104 reservation, DOWN, DRAINED, or not responding.
1105 Nodes which are DOWN, DRAINED, or not responding
1106 will be identified as part of the job's "reason"
1107 field as "UnavailableNodes". Such nodes will typ‐
1108 ically require the intervention of a system
1109 administrator to make available.
1110 Reservation The job is waiting its advanced reservation to
1112 become available.
1113 Resources The job is waiting for resources to become avail‐
1115 able.
1116 SystemFailure Failure of the Slurm system, a file system, the
1118 network, etc.
1119 TimeLimit The job exhausted its time limit.
1121 QOSUsageThreshold Required QOS threshold has been breached.
1123 WaitingForScheduling No reason has been set for this job yet. Waiting
1125 for the scheduler to determine the appropriate
1126 reason.
1127
1130 Jobs typically pass through several states in the course of their exe‐
1131 cution. The typical states are PENDING, RUNNING, SUSPENDED, COMPLET‐
1132 ING, and COMPLETED. An explanation of each state follows.
1133 BF BOOT_FAIL Job terminated due to launch failure, typically due
1135 to a hardware failure (e.g. unable to boot the node
1136 or block and the job can not be requeued).
1137 CA CANCELLED Job was explicitly cancelled by the user or system
1139 administrator. The job may or may not have been
1140 initiated.
1141 CD COMPLETED Job has terminated all processes on all nodes with
1143 an exit code of zero.
1144 CF CONFIGURING Job has been allocated resources, but are waiting
1146 for them to become ready for use (e.g. booting).
1147 CG COMPLETING Job is in the process of completing. Some processes
1149 on some nodes may still be active.
1150 DL DEADLINE Job terminated on deadline.
1152 F FAILED Job terminated with non-zero exit code or other
1154 failure condition.
1155 NF NODE_FAIL Job terminated due to failure of one or more allo‐
1157 cated nodes.
1158 OOM OUT_OF_MEMORY Job experienced out of memory error.
1160 PD PENDING Job is awaiting resource allocation.
1162 PR PREEMPTED Job terminated due to preemption.
1164 R RUNNING Job currently has an allocation.
1166 RD RESV_DEL_HOLD Job is held.
1168 RF REQUEUE_FED Job is being requeued by a federation.
1170 RH REQUEUE_HOLD Held job is being requeued.
1172 RQ REQUEUED Completing job is being requeued.
1174 RS RESIZING Job is about to change size.
1176 RV REVOKED Sibling was removed from cluster due to other clus‐
1178 ter starting the job.
1179 SI SIGNALING Job is being signaled.
1181 SE SPECIAL_EXIT The job was requeued in a special state. This state
1183 can be set by users, typically in EpilogSlurmctld,
1184 if the job has terminated with a particular exit
1185 value.
1186 SO STAGE_OUT Job is staging out files.
1188 ST STOPPED Job has an allocation, but execution has been
1190 stopped with SIGSTOP signal. CPUS have been
1191 retained by this job.
1192 S SUSPENDED Job has an allocation, but execution has been sus‐
1194 pended and CPUs have been released for other jobs.
1195 TO TIMEOUT Job terminated upon reaching its time limit.
1197
1200 Executing squeue sends a remote procedure call to slurmctld. If enough
1201 calls from squeue or other Slurm client commands that send remote pro‐
1202 cedure calls to the slurmctld daemon come in at once, it can result in
1203 a degradation of performance of the slurmctld daemon, possibly result‐
1204 ing in a denial of service.
1205 Do not run squeue or other Slurm client commands that send remote pro‐
1207 cedure calls to slurmctld from loops in shell scripts or other pro‐
1208 grams. Ensure that programs limit calls to squeue to the minimum neces‐
1209 sary for the information you are trying to gather.
1210
1213 Some squeue options may be set via environment variables. These envi‐
1214 ronment variables, along with their corresponding options, are listed
1215 below. (Note: Commandline options will always override these settings.)
1216 SLURM_BITSTR_LEN Specifies the string length to be used for holding
1218 a job array's task ID expression. The default
1219 value is 64 bytes. A value of 0 will print the
1220 full expression with any length required. Larger
1221 values may adversely impact the application perfor‐
1222 mance.
1223 SLURM_CLUSTERS Same as --clusters
1225 SLURM_CONF The location of the Slurm configuration file.
1227 SLURM_TIME_FORMAT Specify the format used to report time stamps. A
1229 value of standard, the default value, generates
1230 output in the form
1231 "year-month-dateThour:minute:second". A value of
1232 relative returns only "hour:minute:second" if the
1233 current day. For other dates in the current year
1234 it prints the "hour:minute" preceded by "Tomorr"
1235 (tomorrow), "Ystday" (yesterday), the name of the
1236 day for the coming week (e.g. "Mon", "Tue", etc.),
1237 otherwise the date (e.g. "25 Apr"). For other
1238 years it returns a date month and year without a
1239 time (e.g. "6 Jun 2012"). All of the time stamps
1240 use a 24 hour format.
1241 A valid strftime() format can also be specified.
1243 For example, a value of "%a %T" will report the day
1244 of the week and a time stamp (e.g. "Mon 12:34:56").
1245 SQUEUE_ACCOUNT -A <account_list>, --account=<account_list>
1247 SQUEUE_ALL -a, --all
1249 SQUEUE_ARRAY -r, --array
1251 SQUEUE_NAMES --name=<name_list>
1253 SQUEUE_FEDERATION --federation
1255 SQUEUE_FORMAT -o <output_format>, --format=<output_format>
1257 SQUEUE_FORMAT2 -O <output_format>, --Format=<output_format>
1259 SQUEUE_LICENSES -p-l <license_list>, --license=<license_list>
1261 SQUEUE_LOCAL --local
1263 SQUEUE_PARTITION -p <part_list>, --partition=<part_list>
1265 SQUEUE_PRIORITY -P, --priority
1267 SQUEUE_QOS -p <qos_list>, --qos=<qos_list>
1269 SQUEUE_SIBLING --sibling
1271 SQUEUE_SORT -S <sort_list>, --sort=<sort_list>
1273 SQUEUE_STATES -t <state_list>, --states=<state_list>
1275 SQUEUE_USERS -u <user_list>, --users=<user_list>
1277
1280 Print the jobs scheduled in the debug partition and in the COMPLETED
1281 state in the format with six right justified digits for the job id fol‐
1282 lowed by the priority with an arbitrary fields size:
1283 # squeue -p debug -t COMPLETED -o "%.6i %p"
1284 JOBID PRIORITY
1285 65543 99993
1286 65544 99992
1287 65545 99991
1288 Print the job steps in the debug partition sorted by user:
1290 # squeue -s -p debug -S u
1291 STEPID NAME PARTITION USER TIME NODELIST
1292 65552.1 test1 debug alice 0:23 dev[1-4]
1293 65562.2 big_run debug bob 0:18 dev22
1294 65550.1 param1 debug candice 1:43:21 dev[6-12]
1295 Print information only about jobs 12345,12345, and 12348:
1297 # squeue --jobs 12345,12346,12348
1298 JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON)
1299 12345 debug job1 dave R 0:21 4 dev[9-12]
1300 12346 debug job2 dave PD 0:00 8 (Resources)
1301 12348 debug job3 ed PD 0:00 4 (Priority)
1302 Print information only about job step 65552.1:
1304 # squeue --steps 65552.1
1305 STEPID NAME PARTITION USER TIME NODELIST
1306 65552.1 test2 debug alice 12:49 dev[1-4]
1307
1310 Copyright (C) 2002-2007 The Regents of the University of California.
1311 Produced at Lawrence Livermore National Laboratory (cf, DISCLAIMER).
1312 Copyright (C) 2008-2010 Lawrence Livermore National Security.
1313 Copyright (C) 2010-2016 SchedMD LLC.
1314 This file is part of Slurm, a resource management program. For
1316 details, see <https://slurm.schedmd.com/>.
1317 Slurm is free software; you can redistribute it and/or modify it under
1319 the terms of the GNU General Public License as published by the Free
1320 Software Foundation; either version 2 of the License, or (at your
1321 option) any later version.
1322 Slurm is distributed in the hope that it will be useful, but WITHOUT
1324 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
1325 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
1326 for more details.
1327
1329 scancel(1), scontrol(1), sinfo(1), srun(1), slurm_load_ctl_conf (3),
1330 slurm_load_jobs (3), slurm_load_node (3), slurm_load_partitions (3)
1331October 2020 Slurm Commands squeue(1)