1slurmd(8) Slurm Daemon slurmd(8)
2
6 slurmd - The compute node daemon for Slurm.
7
10 slurmd [OPTIONS...]
11
14 slurmd is the compute node daemon of Slurm. It monitors all tasks run‐
15 ning on the compute node , accepts work (tasks), launches tasks, and
16 kills running tasks upon request.
17
20 -b Report node rebooted when daemon restarted. Used for testing
21 purposes.
22 -c Clear system locks as needed. This may be required if slurmd
24 terminated abnormally.
25 -C Print the actual hardware configuration (not the configuration
27 from the slurm.conf file) and exit. The format of output is the
28 same as used in slurm.conf to describe a node's configuration
29 plus its uptime.
30 --conf-server <host>[:<port>]
32 Comma-separated list of controllers, the first being the primary
33 slurmctld. A port can (optionally) be specified for each con‐
34 troller. These hosts are where the slurmd will fetch the config‐
35 uration from when running in "configless" mode.
36 -d <file>
38 Specify the fully qualified pathname to the slurmstepd program
39 to be used for shepherding user job steps. This can be useful
40 for testing purposes.
41 -D Run slurmd in the foreground. Error and debug messages will be
43 copied to stderr.
44 -f <file>
46 Read configuration from the specified file. See NOTES below.
47 -F[feature]
49 Start this node as a Dynamic Future node. It will try to match a
50 node definition with a state of FUTURE, optionally using the
51 specified feature to match the node definition.
52 -G Print Generic RESource (GRES) configuration (based upon
54 slurm.conf GRES merged with gres.conf contents for this node)
55 and exit.
56 -h Help; print a brief summary of command options.
58 -L <file>
60 Write log messages to the specified file.
61 -M Lock slurmd pages into system memory using mlockall (2) to dis‐
63 able paging of the slurmd process. This may help in cases where
64 nodes are marked DOWN during periods of heavy swap activity. If
65 the mlockall (2) system call is not available, an error will be
66 printed to the log and slurmd will continue as normal.
67 It is suggested to set LaunchParameters=slurmstepd_memlock in
69 slurm.conf(5) when setting -M.
70 -n <value>
72 Set the daemon's nice value to the specified value, typically a
73 negative number. Also note the PropagatePrioProcess configura‐
74 tion parameter.
75 -N <nodename>
77 Run the daemon with the given nodename. Used to emulate a larger
78 system with more than one slurmd daemon per node. Requires that
79 Slurm be built using the --enable-multiple-slurmd configure op‐
80 tion.
81 -s Change working directory of slurmd to SlurmdLogFile path if pos‐
83 sible, or to SlurmdSpoolDir otherwise. If both of them fail it
84 will fallback to /var/tmp.
85 -v Verbose operation. Multiple -v's increase verbosity.
87 -V, --version
89 Print version information and exit.
90
92 The following environment variables can be used to override settings
93 compiled into slurmd.
94 SLURM_CONF The location of the Slurm configuration file. This
97 is overridden by explicitly naming a configuration
98 file on the command line.
99
101 SIGTERM SIGINT
102 slurmd will shutdown cleanly, waiting for in-progress rollups to
103 finish.
104 SIGHUP Reloads the slurm configuration files, similar to 'scontrol re‐
106 configure'.
107 SIGUSR2
109 Reread the log level from the configs, and then reopen the log
110 file. This should be used when setting up logrotate(8).
111 SIGPIPE
113 This signal is explicitly ignored.
114
116 If slurmd is started with the -D option then the core file will be
117 written to the current working directory. Otherwise if SlurmdLogFile
118 is a fully qualified path name (starting with a slash), the core file
119 will be written to the same directory as the log file. Otherwise the
120 core file will be written to the SlurmSpoolDir directory, or
121 "/var/tmp/" as a last resort. If none of the above directories can be
122 written, no core file will be produced.
123
126 It may be useful to experiment with different slurmd specific configu‐
127 ration parameters using a distinct configuration file (e.g. timeouts).
128 However, this special configuration file will not be used by the slurm‐
129 ctld daemon or the Slurm programs, unless you specifically tell each of
130 them to use it. If you desire changing communication ports, the loca‐
131 tion of the temporary file system, or other parameters used by other
132 Slurm components, change the common configuration file, slurm.conf.
133 If you are using configless mode with a login node that runs a lot of
135 client commands, you may consider running slurmd on that machine so it
136 can manage a cached version of the configuration files. Otherwise, each
137 client command will use the DNS record to contact the controller and
138 get the configuration information, which could place additional load on
139 the controller.
140
143 Copyright (C) 2002-2007 The Regents of the University of California.
144 Copyright (C) 2008-2010 Lawrence Livermore National Security. Copy‐
145 right (C) 2010-2022 SchedMD LLC. Produced at Lawrence Livermore Na‐
146 tional Laboratory (cf, DISCLAIMER).
147 This file is part of Slurm, a resource management program. For de‐
149 tails, see <https://slurm.schedmd.com/>.
150 Slurm is free software; you can redistribute it and/or modify it under
152 the terms of the GNU General Public License as published by the Free
153 Software Foundation; either version 2 of the License, or (at your op‐
154 tion) any later version.
155 Slurm is distributed in the hope that it will be useful, but WITHOUT
157 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
158 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
159 for more details.
160
163 /etc/slurm.conf
164
167 slurm.conf(5), slurmctld(8)
168May 2022 Slurm Daemon slurmd(8)