slurmctld(8)

1slurmctld(8)                     Slurm Daemon                     slurmctld(8)
2
3
4

NAME

6       slurmctld - The central management daemon of Slurm.
7

SYNOPSIS

9       slurmctld [OPTIONS...]
10

DESCRIPTION

12       slurmctld  is  the  central management daemon of Slurm. It monitors all
13       other Slurm daemons and resources, accepts work (jobs),  and  allocates
14       resources to those jobs. Given the critical functionality of slurmctld,
15       there may be a backup server to assume these  functions  in  the  event
16       that the primary server fails.
17
18       OPTIONS
19
20
21       -c     Clear  all  previous  slurmctld  state from its last checkpoint.
22              With this option, all jobs, including both running  and  queued,
23              and all node states, will be deleted.  Without this option, pre‐
24              viously running jobs will be preserved along with node State  of
25              DOWN, DRAINED and DRAINING nodes and the associated Reason field
26              for those nodes.  NOTE: It is rare you would ever  want  to  use
27              this in production as all jobs will be killed.
28
29
30       -d     Run slurmctld in the background.
31
32       -D     Run slurmctld in the foreground with logging copied to stdout.
33
34       -f <file>
35              Read configuration from the specified file. See NOTES below.
36
37       -h     Help; print a brief summary of command options.
38
39       -i     Ignore errors found while reading in state files on startup.
40
41       -L <file>
42              Write log messages to the specified file.
43
44
45       -n <value>
46              Set  the daemon's nice value to the specified value, typically a
47              negative number.
48
49
50       -r     Recover partial  state  from  last  checkpoint:  jobs  and  node
51              DOWN/DRAIN  state  and  reason  information state.  No partition
52              state is recovered.  This is the default action.
53
54
55       -R     Recover full state from last checkpoint: jobs, node, and  parti‐
56              tion  state.   Without this option, previously running jobs will
57              be preserved along with node State of DOWN, DRAINED and DRAINING
58              nodes  and the associated Reason field for those nodes. No other
59              node or partition state will be preserved.
60
61
62       -v     Verbose operation. Multiple -v's increase verbosity.
63
64       -V     Print version information and exit.
65
66

ENVIRONMENT VARIABLES

68       The following environment variables can be used  to  override  settings
69       compiled into slurmctld.
70
71       SLURM_CONF          The  location of the Slurm configuration file. This
72                           is overridden by explicitly naming a  configuration
73                           file on the command line.
74
75

CORE FILE LOCATION

77       If  slurmctld  is started with the -D option then the core file will be
78       written to the current working directory.  Otherwise  if  SlurmctldLog‐
79       File  is  a fully qualified path name (starting with a slash), the core
80       file will be written to the same directory as the  log  file,  provided
81       SlurmUser  has  write  permission on the directory.  Otherwise the core
82       file will be written to the StateSaveLocation, or "/var/tmp/" as a last
83       resort.  If  none  of  the  above directories have write permission for
84       SlurmUser, no core file will be produced.  The command "scontrol abort"
85       can be used to abort the slurmctld daemon and generate a core file.
86
87

NOTES

89       It  may  be useful to experiment with different slurmctld specific con‐
90       figuration parameters using a distinct configuration file  (e.g.  time‐
91       outs).   However,  this  special configuration file will not be used by
92       the slurmd daemon or the Slurm programs, unless you  specifically  tell
93       each of them to use it. If you desire changing communication ports, the
94       location of the temporary file system,  or  other  parameters  used  by
95       other   Slurm   components,   change  the  common  configuration  file,
96       slurm.conf.
97
98

COPYING

100       Copyright (C) 2002-2007 The Regents of the  University  of  California.
101       Copyright (C) 2008-2010 Lawrence Livermore National Security.  Produced
102       at   Lawrence   Livermore   National   Laboratory   (cf,   DISCLAIMER).
103       CODE-OCEC-09-009. All rights reserved.
104
105       This  file  is  part  of  Slurm,  a  resource  management program.  For
106       details, see <https://slurm.schedmd.com/>.
107
108       Slurm is free software; you can redistribute it and/or modify it  under
109       the  terms  of  the GNU General Public License as published by the Free
110       Software Foundation; either version 2  of  the  License,  or  (at  your
111       option) any later version.
112
113       Slurm  is  distributed  in the hope that it will be useful, but WITHOUT
114       ANY WARRANTY; without even the implied warranty of  MERCHANTABILITY  or
115       FITNESS  FOR  A PARTICULAR PURPOSE.  See the GNU General Public License
116       for more details.
117
118