1smap(1) Slurm Commands smap(1)
2
6 smap - graphically view information about Slurm jobs, partitions, and
7 set configurations parameters.
8
11 smap [OPTIONS...]
12
14 smap is used to graphically view job, partition and node information
15 for a system running Slurm. Note that information about nodes and par‐
16 titions to which you lack access will always be displayed to avoid
17 obvious gaps in the output. This is equivalent to the --all option of
18 the sinfo and squeue commands.
19
22 -c, --commandline
23 Print output to the commandline, no curses.
24 -D <option>, --display=<option>
27 sets the display mode for smap, showing relevant information
28 about the selected view and displaying a corresponding node
29 chart. Note that unallocated nodes are indicated by a '.' and
30 nodes in the DOWN, DRAINED or FAIL state by a '#'. When the
31 --iterate=<seconds> option is also selected, you can switch dis‐
32 plays by typing a different letter from the list below.
33 j Displays information about jobs running on system.
35 r Display information about advanced reservations. While
37 all current and future reservations will be listed, only
38 currently active reservations will appear on the node
39 map.
40 s Displays information about slurm partitions on the system
42 -h, --noheader
45 Do not print a header on the output.
46 -H, --show_hidden
49 Display hidden partitions and their jobs.
50 --help,
53 Print a message describing all smap options.
54 -i <seconds> , --iterate=<seconds>
57 Print the state on a periodic basis. Sleep for the indicated
58 number of seconds between reports. User can exit at anytime by
59 typing 'q' or hitting the return key. If user is in configure
60 mode type 'exit' to exit program, 'quit' to exit configure mode.
61 -M, --clusters=<string>
64 Clusters to issue commands to. Note that the SlurmDBD must be
65 up for this option to work properly.
66 -n, --nodes
69 Only show objects with these nodes.
70 -Q, --quiet
73 Avoid printing error messages.
74 --usage
77 Print a brief message listing the smap options.
78 -V , --version
81 Print version information and exit.
82
85 When using smap in curses mode and when the --iterate=<seconds> option
86 is also selected, you can scroll through the different windows using
87 the arrow keys. The up and down arrow keys scroll the window contain‐
88 ing the grid, and the left and right arrow keys scroll the window con‐
89 taining the text information.
90 With the iterate option selected, you can use any of the options avail‐
92 able to the -D option listed above (except 'c') to change screens. You
93 can also hide or make visible hidden partitions by pressing 'h' at any
94 moment.
95
98 ACCESS_CONTROL
99 Identifies the users or bank accounts which can use this
100 advanced reservation. A prefix of "A:" indicates that the fol‐
101 lowing account names may use this reservation. A prefix of "U:"
102 indicates that the following user names may use this reserva‐
103 tion.
104 AVAIL Partition state: up or down.
106 END_TIME
108 The time when an advanced reservation ended.
109 ID Key to identify the nodes associated with this entity in the
111 node chart.
112 NAME Name of the job or advanced reservation.
114 NODELIST
116 Names of nodes associated with this configuration, partition or
117 reservation.
118 NODES Count of nodes with this particular configuration.
120 PARTITION
122 Name of a partition. Note that the suffix "*" identifies the
123 default partition.
124 ST State of a job in compact form. Possible states include: PD
126 (pending), R (running), S (suspended), CD (completed), CF (con‐
127 figuring), CG (completing), F (failed), TO (timeout), and NF
128 (node failure). See JOB STATE CODES section below for more
129 information.
130 START_TIME
132 The time when an advanced reservation started.
133 STATE State of the nodes. Possible states include: allocated, com‐
135 pleting, down, drained, draining, fail, failing, idle, and
136 unknown plus their abbreviated forms: alloc, comp, down, drain,
137 drng, fail, failg, idle, and unk respectively. Note that the
138 suffix "*" identifies nodes that are presently not responding.
139 See NODE STATE CODES section below for more information.
140 TIMELIMIT
142 Maximum time limit for any user job in days-hours:minutes:sec‐
143 onds. infinite is used to identify jobs or partitions without a
144 job time limit.
145 TOPOGRAPHY INFORMATION
149 The node chart is designed to indicate relative locations of the nodes.
151 On most Linux clusters this will represent a one-dimensional array of
152 nodes. Larger clusters will utilize multiple as needed with right side
153 of one line being logically followed by the left side of the next line.
154
157 Node state codes are shortened as required for the field size. These
158 node states may be followed by a special character to identify state
159 flags associated with the node. The following node sufficies and
160 states are used:
161 * The node is presently not responding and will not be allocated any
163 new work. If the node remains non-responsive, it will be placed in
164 the DOWN state (except in the case of COMPLETING, DRAINED, DRAIN‐
165 ING, FAIL, FAILING nodes).
166 ~ The node is presently in a power saving mode (typically running at
168 reduced frequency).
169 # The node is presently being powered up or configured.
171 $ The node is currently in a reservation with a flag value of "main‐
173 tenance".
174 @ The node is pending reboot.
176 ALLOCATED The node has been allocated to one or more jobs.
178 ALLOCATED+ The node is allocated to one or more active jobs plus one
180 or more jobs are in the process of COMPLETING.
181 COMPLETING All jobs associated with this node are in the process of
183 COMPLETING. This node state will be removed when all of
184 the job's processes have terminated and the Slurm epilog
185 program (if any) has terminated. See the Epilog parameter
186 description in the slurm.conf man page for more informa‐
187 tion.
188 DOWN The node is unavailable for use. Slurm can automatically
190 place nodes in this state if some failure occurs. System
191 administrators may also explicitly place nodes in this
192 state. If a node resumes normal operation, Slurm can auto‐
193 matically return it to service. See the ReturnToService and
194 SlurmdTimeout parameter descriptions in the slurm.conf(5)
195 man page for more information.
196 DRAINED The node is unavailable for use per system administrator
198 request. See the update node command in the scontrol(1)
199 man page or the slurm.conf(5) man page for more informa‐
200 tion.
201 DRAINING The node is currently executing a job, but will not be
203 allocated to additional jobs. The node state will be
204 changed to state DRAINED when the last job on it completes.
205 Nodes enter this state per system administrator request.
206 See the update node command in the scontrol(1) man page or
207 the slurm.conf(5) man page for more information.
208 FAIL The node is expected to fail soon and is unavailable for
210 use per system administrator request. See the update node
211 command in the scontrol(1) man page or the slurm.conf(5)
212 man page for more information.
213 FAILING The node is currently executing a job, but is expected to
215 fail soon and is unavailable for use per system administra‐
216 tor request. See the update node command in the scon‐
217 trol(1) man page or the slurm.conf(5) man page for more
218 information.
219 IDLE The node is not allocated to any jobs and is available for
221 use.
222 MAINT The node is currently in a reservation with a flag value of
224 "maintainence".
225 REBOOT The node is currently scheduled to be rebooted.
227 UNKNOWN The Slurm controller has just started and the node's state
229 has not yet been determined.
230
233 Jobs typically pass through several states in the course of their exe‐
234 cution. The typical states are PENDING, RUNNING, SUSPENDED, COMPLET‐
235 ING, and COMPLETED. An explanation of each state follows.
236 BF BOOT_FAIL Job terminated due to launch failure, typically due
238 to a hardware failure (e.g. unable to boot the node
239 or block and the job can not be requeued).
240 CA CANCELLED Job was explicitly cancelled by the user or system
242 administrator. The job may or may not have been
243 initiated.
244 CD COMPLETED Job has terminated all processes on all nodes with
246 an exit code of zero.
247 CG COMPLETING Job is in the process of completing. Some processes
249 on some nodes may still be active.
250 CF CONFIGURING Job has been allocated resources, but are waiting
252 for them to become ready for use (e.g. booting).
253 F FAILED Job terminated with non-zero exit code or other
255 failure condition.
256 NF NODE_FAIL Job terminated due to failure of one or more allo‐
258 cated nodes.
259 PD PENDING Sibling job (in federation) revoked.
261 PR PREEMPTED Job terminated due to preemption.
263 RV REVOKED Job currently has an allocation.
265 R RUNNING Job currently has an allocation.
267 SI SIGNALING Signal of job currently in progress.
269 SO STAGE_OUT Staging out data after job completion.
271 SE SPECIAL_EXIT The job was requeued in a special state. This state
273 can be set by users, typically in EpilogSlurmctld,
274 if the job has terminated with a particular exit
275 value.
276 ST STOPPED Job has an allocation, but execution has been
278 stopped with SIGSTOP signal. CPUS have been
279 retained by this job.
280 S SUSPENDED Job has an allocation, but execution has been sus‐
282 pended and CPUs have been released for other jobs.
283 TO TIMEOUT Job terminated upon reaching its time limit.
285
288 The following environment variables can be used to override settings
289 compiled into smap.
290 SLURM_CONF The location of the Slurm configuration file.
292
295 Copyright (C) 2004-2007 The Regents of the University of California.
296 Produced at Lawrence Livermore National Laboratory (cf, DISCLAIMER).
297 Copyright (C) 2008-2009 Lawrence Livermore National Security.
298 Copyright (C) 2010-2013 SchedMD LLC.
299 This file is part of Slurm, a resource management program. For
301 details, see <https://slurm.schedmd.com/>.
302 Slurm is free software; you can redistribute it and/or modify it under
304 the terms of the GNU General Public License as published by the Free
305 Software Foundation; either version 2 of the License, or (at your
306 option) any later version.
307 Slurm is distributed in the hope that it will be useful, but WITHOUT
309 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
310 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
311 for more details.
312
315 scontrol(1), sinfo(1), squeue(1), slurm_load_ctl_conf [22m(3),
316 slurm_load_jobs [22m(3), slurm_load_node [22m(3), slurm_load_partitions (3),
317 slurm_reconfigure [22m(3), slurm_shutdown [22m(3), slurm_update_job [22m(3),
318 slurm_update_node (3), slurm_update_partition (3), slurm.conf(5)
319March 2018 Slurm Commands smap(1)