update: Allow passing command line arguments to task now.
1.What is clojure-control?
The idea came from node-control.
Define clusters and tasks for system administration or code deployment, then execute them on one or many remote machines.
Clojure-control depends only on OpenSSH and clojure on the local
control machine.Remote machines simply need a standard sshd daemon.
2.Quick example
Get the current date from the two machines listed in the 'mycluster' config with a single command:
(ns samples
(:use [control.core :only [task cluster scp ssh begin]]))
;;define clusters
(cluster :mycluster
:clients [
{ :host "a.domain.com" :user "alogin"}
{ :host "b.domain.com" :user "blogin"}
])
;;define tasks
(task :date "Get date"
[]
(ssh "date"))
;;start running
(begin)
If saved in a file named "controls.clj",run with
java -cp clojure.jar:clojure-contrib.jar:control-0.1-SNAPSHOT.jar clojure.main controls.clj mycluster date
Each machine execute "date" command ,and the output form the remote
machine is printed to the console.Exmaple console output
Performing mycluster
Performing date for a.domain.com
a.domain.com:ssh: date
a.domain.com:stdout: Sun Jul 24 19:14:09 CST 2011
a.domain.com:exit: 0
Performing date for b.domain.com
b.domain.com:ssh: date
b.domain.com:stdout: Sun Jul 24 19:14:09 CST 2011
b.domain.com:exit: 0
Each line of output is labeled with the address of the machine the
command was executed on. The actual command sent and the user used to
send it is displayed. stdout and stderr output of the remote process
is identified as well as the final exit code of the local ssh
command.
3.How to scp files?
Let's define a new task named deploy
(task :deploy "scp files to remote machines"
[]
(scp ("release1.tar.gz" "release2.tar.gz") "/home/alogin/"))
Then it will copy release1.tar.gz and release2.tar.gz to remote
machine's /home/alogin directory.
4.Where is it?
It's on github,
https://github.com/killme2008/clojure-control Any suggestion or bug reports welcomed.