1CLUSH(1) ClusterShell User Manual CLUSH(1)
2
6 clush - execute shell commands on a cluster
7
9 clush -a | -g group | -w nodes [OPTIONS]
10 clush -a | -g group | -w nodes [OPTIONS] command
12 clush -a | -g group | -w nodes [OPTIONS] --copy file | dir [ file |
14 dir ...] [ --dest path ]
15 clush -a | -g group | -w nodes [OPTIONS] --rcopy file | dir [ file |
17 dir ...] [ --dest path ]
18
20 clush is a program for executing commands in parallel on a cluster and
21 for gathering their results. clush executes commands interactively or
22 can be used within shell scripts and other applications. It is a par‐
23 tial front-end to the ClusterShell library that ensures a light, uni‐
24 fied and robust parallel command execution framework. Thus, it allows
25 traditional shell scripts to benefit from some of the library features.
26 clush currently makes use of the Ssh worker of ClusterShell, by de‐
27 fault, that only requires ssh(1) (OpenSSH SSH client).
28
30 clush can be started non-interactively to run a shell command, or can
31 be invoked as an interactive shell. To start a clush interactive ses‐
32 sion, invoke the clush command without providing command.
33 Non-interactive mode
35 When clush is started non-interactively, the command is executed
36 on the specified remote hosts in parallel. If option -b or
37 --dshbak is specified, clush waits for command completion and
38 then displays gathered output results.
39 The -w option allows you to specify remote hosts by using Clus‐
41 terShell NodeSet syntax, including the node groups @group spe‐
42 cial syntax and the Extended Patterns syntax to benefits from
43 NodeSet basic arithmetic (like @Agroup\&@Bgroup). See EXTENDED
44 PATTERNS in nodeset(1) and also groups.conf(5) for more informa‐
45 tion.
46 Unless the option --nostdin (or -n) is specified, clush detects
48 when its standard input is connected to a terminal (as deter‐
49 mined by isatty(3)). If actually connected to a terminal, clush
50 listens to standard input when commands are running, waiting for
51 an Enter key press. Doing so will display the status of current
52 nodes. If standard input is not connected to a terminal, and
53 unless the option --nostdin is specified, clush binds the stan‐
54 dard input of the remote commands to its own standard input, al‐
55 lowing scripting methods like:
56 # echo foo | clush -w node[40-42] -b cat
57 ---------------
58 node[40-42]
59 ---------------
60 foo
61 Please see some other great examples in the EXAMPLES section be‐
64 low.
65 Interactive session
67 If a command is not specified, and its standard input is con‐
68 nected to a terminal, clush runs interactively. In this mode,
69 clush uses the GNU readline library to read command lines. Read‐
70 line provides commands for searching through the command history
71 for lines containing a specified string. For instance, type Con‐
72 trol-R to search in the history for the next entry matching the
73 search string typed so far. clush also recognizes special sin‐
74 gle-character prefixes that allows the user to see and modify
75 the current nodeset (the nodes where the commands are executed).
76 Single-character interactive commands are:
78 clush> ?
80 show current nodeset
81 clush> @<NODESET>
83 set current nodeset
84 clush> +<NODESET>
86 add nodes to current nodeset
87 clush> -<NODESET>
89 remove nodes from current nodeset
90 clush> !COMMAND
92 execute COMMAND on the local system
93 clush> =
95 toggle the output format (gathered or standard
96 mode)
97 To leave an interactive session, type quit or Control-D.
99 Local execution ( --worker=exec or -R exec )
101 Instead of running provided command on remote nodes, clush can
102 use the dedicated exec worker to launch the command locally, for
103 each node. Some parameters could be used in the command line to
104 make a different command for each node. %h or %host will be re‐
105 placed by node name and %n or %rank by the remote rank [0-N] (to
106 get a literal % use %%)
107 File copying mode ( --copy )
109 When clush is started with the -c or --copy option, it will at‐
110 tempt to copy specified file and/or dir to the provided target
111 cluster nodes. If the --dest option is specified, it will put
112 the copied files there.
113 Reverse file copying mode ( --rcopy )
115 When clush is started with the --rcopy option, it will attempt
116 to retrieve specified file and/or dir from provided cluster
117 nodes. If the --dest option is specified, it must be a directory
118 path where the files will be stored with their hostname ap‐
119 pended. If the destination path is not specified, it will take
120 the first file or dir basename directory as the local destina‐
121 tion.
122
124 --version
125 show clush version number and exit
126 -s GROUPSOURCE, --groupsource=GROUPSOURCE
128 optional groups.conf(5) group source to use
129 -n, --nostdin
131 do not watch for possible input from stdin; this should be used
132 when clush is run in the background (or in scripts).
133 --groupsconf=FILE
135 use alternate config file for groups.conf(5)
136 --conf=FILE
138 use alternate config file for clush.conf(5)
139 -O <KEY=VALUE>, --option=<KEY=VALUE>
141 override any key=value clush.conf(5) options (repeat as needed)
142 Selecting target nodes:
144 -w NODES
146 nodes where to run the command
147 -x NODES
149 exclude nodes from the node list
150 -a, --all
152 run command on all nodes
153 -g GROUP, --group=GROUP
155 run command on a group of nodes
156 -X GROUP
158 exclude nodes from this group
159 --hostfile=FILE, --machinefile=FILE
161 path to a file containing a list of single hosts, node
162 sets or node groups, separated by spaces and lines (may
163 be specified multiple times, one per file)
164 --topology=FILE
166 topology configuration file to use for tree mode
167 --pick=N
169 pick N node(s) at random in nodeset
170 Output behaviour:
172 -q, --quiet
174 be quiet, print essential output only
175 -v, --verbose
177 be verbose, print informative messages
178 -d, --debug
180 output more messages for debugging purpose
181 -G, --groupbase
183 do not display group source prefix
184 -L disable header block and order output by nodes; if -b/-B
186 is not specified, clush will wait for all commands to
187 finish and then display aggregated output of commands
188 with same return codes, ordered by node name; alterna‐
189 tively, when used in conjunction with -b/-B (eg. -bL),
190 clush will enable a "life gathering" of results by line,
191 such as the next line is displayed as soon as possible
192 (eg. when all nodes have sent the line)
193 -N disable labeling of command line
195 -P, --progress
197 show progress during command execution; if writing is
198 performed to standard input, the live progress indicator
199 will display the global bandwidth of data written to the
200 target nodes
201 -b, --dshbak
203 display gathered results in a dshbak-like way (note: it
204 will only try to aggregate the output of commands with
205 same return codes)
206 -B like -b but including standard error
208 -r, --regroup
210 fold nodeset using node groups
211 -S, --maxrc
213 return the largest of command return codes
214 --color=WHENCOLOR
216 clush can use NO_COLOR, CLICOLOR and CLICOLOR_FORCE envi‐
217 ronment variables. NO_COLOR takes precedence over CLI‐
218 COLOR_FORCE which takes precedence over CLICOLOR. When
219 --color option is used these environment variables are
220 not taken into account. --color tells whether to use ANSI
221 colors to surround node or nodeset prefix/header with es‐
222 cape sequences to display them in color on the terminal.
223 WHENCOLOR is never, always or auto (which use color if
224 standard output/error refer to a terminal). Colors are
225 set to [34m (blue foreground text) for stdout and [31m
226 (red foreground text) for stderr, and cannot be modified.
227 --diff show diff between common outputs (find the best reference
229 output by focusing on largest nodeset and also smaller
230 command return code)
231 File copying:
233 -c, --copy
235 copy local file or directory to remote nodes
236 --rcopy
238 copy file or directory from remote nodes
239 --dest=DEST_PATH
241 destination file or directory on the nodes (optional: use
242 the first source directory path when not specified)
243 -p preserve modification times and modes
245 Connection options:
247 -f FANOUT, --fanout=FANOUT
249 do not execute more than FANOUT commands at the same
250 time, useful to limit resource usage. In tree mode, the
251 same fanout value is used on the head node and on each
252 gateway (the fanout value is propagated). That is, if the
253 fanout is 16, each gateway will initate up to 16 connec‐
254 tions to their target nodes at the same time. Default
255 fanout value is defined in clush.conf(5).
256 -l USER, --user=USER
258 execute remote command as user
259 -o OPTIONS, --options=OPTIONS
261 can be used to give ssh options, eg. -o "-p 2022 -i
262 ~/.ssh/myidrsa"; these options are added first to ssh and
263 override default ones
264 -t CONNECT_TIMEOUT, --connect_timeout=CONNECT_TIMEOUT
266 limit time to connect to a node
267 -u COMMAND_TIMEOUT, --command_timeout=COMMAND_TIMEOUT
269 limit time for command to run on the node
270 -R WORKER, --worker=WORKER
272 worker name to use for connection (exec, ssh, rsh, pdsh,
273 or the name of a Python worker module), default is ssh
274 --remote=REMOTE
276 whether to enable remote execution: in tree mode, 'yes'
277 forces connections to the leaf nodes for execution, 'no'
278 establishes connections up to the leaf parent nodes for
279 execution (default is 'yes')
280 For a short explanation of these options, see -h, --help.
282
284 By default, an exit status of zero indicates success of the clush com‐
285 mand but gives no information about the remote commands exit status.
286 However, when the -S option is specified, the exit status of clush is
287 the largest value of the remote commands return codes.
288 For failed remote commands whose exit status is non-zero, and unless
290 the combination of options -qS is specified, clush displays messages
291 similar to:
292 clush: node[40-42]: exited with exit code 1
294
296 Remote parallel execution
297 # clush -w node[3-5,62] uname -r
298 Run command uname -r in parallel on nodes: node3, node4, node5
299 and node62
300 Local parallel execution
302 # clush -w node[1-3] --worker=exec ping -c1 %host
303 Run locally, in parallel, a ping command for nodes: node1, node2
304 and node3. You may also use -R exec as the shorter and pdsh
305 compatible option.
306 Display features
308 # clush -w node[3-5,62] -b uname -r
309 Run command uname -r on nodes[3-5,62] and display gathered out‐
310 put results (integrated dshbak-like).
311 # clush -w node[3-5,62] -bL uname -r
313 Line mode: run command uname -r on nodes[3-5,62] and display
314 gathered output results without default header block.
315 # ssh node32 find /etc/yum.repos.d -type f | clush -w node[40-42] -b
317 xargs ls -l
318 Search some files on node32 in /etc/yum.repos.d and use clush to
319 list the matching ones on node[40-42], and use -b to display
320 gathered results.
321 # clush -w node[3-5,62] --diff dmidecode -s bios-version
323 Run this Linux command to get BIOS version on nodes[3-5,62] and
324 show version differences (if any).
325 All nodes
327 # clush -a uname -r
328 Run command uname -r on all cluster nodes, see groups.conf(5) to
329 setup all cluster nodes (all: field).
330 # clush -a -x node[5,7] uname -r
332 Run command uname -r on all cluster nodes except on nodes node5
333 and node7.
334 # clush -a --diff cat /some/file
336 Run command cat /some/file on all cluster nodes and show differ‐
337 ences (if any), line by line, between common outputs.
338 Node groups
340 # clush -w @oss modprobe lustre
341 Run command modprobe lustre on nodes from node group named oss,
342 see groups.conf(5) to setup node groups (map: field).
343 # clush -g oss modprobe lustre
345 Same as previous example but using -g to avoid @ group prefix.
346 # clush -w @mds,@oss modprobe lustre
348 You may specify several node groups by separating them with com‐
349 mas (please see EXTENDED PATTERNS in nodeset(1) and also
350 groups.conf(5) for more information).
351 Copy files
353 # clush -w node[3-5,62] --copy /etc/motd
354 Copy local file /etc/motd to remote nodes node[3-5,62].
355 # clush -w node[3-5,62] --copy /etc/motd --dest /tmp/motd2
357 Copy local file /etc/motd to remote nodes node[3-5,62] at path
358 /tmp/motd2.
359 # clush -w node[3-5,62] -c /usr/share/doc/clustershell
361 Recursively copy local directory /usr/share/doc/clustershell to
362 the same path on remote nodes node[3-5,62].
363 # clush -w node[3-5,62] --rcopy /etc/motd --dest /tmp
365 Copy /etc/motd from remote nodes node[3-5,62] to local /tmp di‐
366 rectory, each file having their remote hostname appended, eg.
367 /tmp/motd.node3.
368
370 /etc/clustershell/clush.conf
371 System-wide clush configuration file.
372 $XDG_CONFIG_HOME/clustershell/clush.conf
374 User configuration file for clush. If $XDG_CONFIG_HOME is not
375 defined, $HOME/.config/clustershell/clush.conf is used instead.
376 $HOME/.local/etc/clustershell/clush.conf
378 Local user configuration file for clush (default installation
379 for pip --user)
380 ~/.clush.conf
382 Deprecated per-user clush configuration file.
383 ~/.clush_history
385 File in which interactive clush command history is saved.
386
388 clubak(1), cluset(1), nodeset(1), readline(3), clush.conf(5),
389 groups.conf(5).
390 http://clustershell.readthedocs.org/
392
394 Use the following URL to submit a bug report or feedback:
395 https://github.com/cea-hpc/clustershell/issues
396
398 Stephane Thiell <sthiell@stanford.edu>
399
401 GNU Lesser General Public License version 2.1 or later (LGPLv2.1+)
4021.8.4 2021-11-03 CLUSH(1)