Linux service --系统服务管理

用途说明

service命令用于对系统服务进行管理，比如启动（start）、停止（stop）、重启（restart）、查看状态（status）等。相关的命令还包括chkconfig、ntsysv等，chkconfig用于查看、设置服务的运行级别，ntsysv用于直观方便的设置各个服务是否自动启动。service命令本身是一个shell脚本，它在/etc/init.d/目录查找指定的服务脚本，然后调用该服务脚本来完成任务。

看看下面的手册页可能更加清楚的了解service的内幕：service运行指定服务（称之为SystemV初始脚本）时，把大部分环境变量去掉了，只保留LANG和TERM两个环境变量，并且把当前路径置为/，也就是说是在一个可以预测的非常干净的环境中运行服务脚本。这种脚本保存在/etc/init.d目录中，它至少要支持start和stop命令。

 

man service 写道

service(8) service(8)

 

NAME

service - run aSystem V init script

 

SYNOPSIS

service SCRIPTCOMMAND [OPTIONS]

 

service --status-all

 

service --help | -h |--version

 

DESCRIPTION

service runs a SystemV init script in as predictable environment as possible, removing mostenvironment vari-

ables and withcurrent working directory set to /.

 

The SCRIPT parameterspecifies a System V init script, located in /etc/init.d/SCRIPT. The supportedvalues of

COMMAND depend on theinvoked script, service passes COMMAND and OPTIONS it to the init scriptunmodified. All

scripts shouldsupport at least the start and stop commands. As a special case, if COMMAND is--full-restart,

the script is runtwice, first with the stop command, then with the start command.

 

service --status-allruns all init scripts, in alphabetical order, with the status command.

 

FILES

/etc/init.d

The directorycontaining System V init scripts.

 

ENVIRONMENT

LANG, TERM

The only environmentvariables passed to the init scripts.

 

SEE ALSO

chkconfig(8),ntsysv(8)

 

Jan 2006 service(8)

 

常用方式

格式：service<service>

打印指定服务<service>的命令行使用帮助。

 

格式：service<service> start

启动指定的系统服务<service>

 

格式：service<service> stop

停止指定的系统服务<service>

 

格式：service<service> restart

重新启动指定的系统服务<service>，即先停止（stop），然后再启动（start）。

 

格式：chkconfig --list

查看系统服务列表，以及每个服务的运行级别。

 

格式：chkconfig<service> on

设置指定服务<service>开机时自动启动。

 

格式：chkconfig<service> off

设置指定服务<service>开机时不自动启动。

 

格式：ntsysv

以全屏幕文本界面设置服务开机时是否自动启动。

 

使用示例

示例一 网络重启

当修改了主机名、ip地址等信息时，经常需要把网络重启使之生效。

[root@node34 root]# service network

用法：/etc/init.d/network{start|stop|restart|reload|status}

[root@node34 root]# service network status

配置设备：

lo eth0

当前的活跃设备：

lo eth0

[root@node34 root]# service network restart

正在关闭接口eth0：                                       [  确定  ]

关闭环回接口：                                            [  确定  ]

设置网络参数：                                            [  确定  ]

弹出环回接口：                                            [  确定  ]

弹出界面eth0：                                           [  确定  ]

[root@node34 root]#

 

示例二 重启MySQL

[root@node34 root]# service mysql

mysql: unrecognizedservice

[root@node34 root]# service mysqld

用法：/etc/init.d/mysqld{start|stop|status|condrestart|restart}

[root@node34 root]# service mysqld status

mysqld (pid 1638)正在运行...

[root@node34 root]# service mysqld restart

停止MySQL：                                              [  确定  ]

启动MySQL：                                              [  确定  ]

[root@node34 root]#

 

示例三 service脚本源码展示

[root@web ~]# cat /sbin/service

#!/bin/sh

 

./etc/init.d/functions

 

VERSION="`basename$0` ver. 0.91"

USAGE="Usage:`basename $0` < option > | --status-all | \

[ service_name [command | --full-restart ] ]"

SERVICE=

SERVICEDIR="/etc/init.d"

OPTIONS=

 

if [ $# -eq 0 ]; then

  echo "${USAGE}" >&2

  exit 1

fi

 

cd /

while [ $# -gt 0 ];do

 case "${1}" in

   --help | -h | --h* )

      echo "${USAGE}" >&2

      exit 0

      ;;

   --version | -V )

      echo "${VERSION}" >&2

      exit 0

      ;;

   *)

      if [ -z "${SERVICE}" -a $# -eq 1 -a "${1}" ="--status-all" ]; then

         cd ${SERVICEDIR}

         for SERVICE in * ; do

           case "${SERVICE}" in

             functions | halt | killall | single| linuxconf| kudzu)

                 ;;

             *)

               if ! is_ignored_file "${SERVICE}" \

                   && [ -x "${SERVICEDIR}/${SERVICE}" ]; then

                 env -i LANG="$LANG" PATH="$PATH" TERM="$TERM""${SERVICEDIR}/${SERVICE}" status

               fi

               ;;

           esac

         done

         exit 0

      elif [ $# -eq 2 -a "${2}" = "--full-restart" ]; then

         SERVICE="${1}"

         if [ -x "${SERVICEDIR}/${SERVICE}" ]; then

           env -i LANG="$LANG" PATH="$PATH" TERM="$TERM""${SERVICEDIR}/${SERVICE}" stop

           env -i LANG="$LANG" PATH="$PATH" TERM="$TERM""${SERVICEDIR}/${SERVICE}" start

           exit $?

         fi

      elif [ -z "${SERVICE}" ]; then

        SERVICE="${1}"

      else

        OPTIONS="${OPTIONS} ${1}"

      fi

      shift

      ;;

  esac

done

 

if [ -x"${SERVICEDIR}/${SERVICE}" ]; then

  env -i LANG="$LANG"PATH="$PATH" TERM="$TERM""${SERVICEDIR}/${SERVICE}" ${OPTIONS}

else

  echo $"${SERVICE}: unrecognized service" >&2

  exit 1

fi

[root@web ~]#

 

示例四 crond服务的源码

[root@web init.d]# cat /etc/init.d/crond

#! /bin/bash

#

#crond         Start/Stop the cron clock daemon.

#

# chkconfig: 2345 90 60

# description: cronis a standard UNIX program that runs user-specified \

#             programs at periodic scheduled times. vixie cron adds a \

#             number of features to the basic UNIX cron, including better \

#             security and more powerful configuration options.

# processname: crond

# config:/etc/crontab

# pidfile:/var/run/crond.pid

 

# Source functionlibrary.

./etc/init.d/functions

./etc/sysconfig/crond

t=${CRON_VALIDATE_MAILRCPTS:-UNSET}

[ "$t" !="UNSET" ] && export CRON_VALIDATE_MAILRCPTS="$t"

 

# See how we werecalled.

 

prog="crond"

 

start() {

       echo -n $"Starting $prog: "

       if [ -e /var/lock/subsys/crond ]; then

           if [ -e /var/run/crond.pid ] && [ -e /proc/`cat /var/run/crond.pid` ];then

               echo -n $"cannot start crond: crond is already running.";

               failure $"cannot start crond: crond already running.";

               echo

               return 1

           fi

       fi

       daemon crond $CRONDARGS

       RETVAL=$?

       echo

       [ $RETVAL -eq 0 ] && touch /var/lock/subsys/crond;

       return $RETVAL

}

 

stop() {

       echo -n $"Stopping $prog: "

       if [ ! -e /var/lock/subsys/crond ]; then

           echo -n $"cannot stop crond: crond is not running."

           failure $"cannot stop crond: crond is not running."

           echo

           return 1;

       fi

       killproc crond

       RETVAL=$?

       echo

       [ $RETVAL -eq 0 ] && rm -f /var/lock/subsys/crond;

       return $RETVAL

}

 

rhstatus() {

       status crond

}

 

restart() {

       stop

       start

}

 

reload() {

       echo -n $"Reloading cron daemon configuration: "

       killproc crond -HUP

       RETVAL=$?

       echo

       return $RETVAL

}

 

case "$1"in

 start)

       start

       ;;

 stop)

       stop

       ;;

 restart)

       restart

       ;;

 reload)

       reload

       ;;

 status)

       rhstatus

       ;;

 condrestart)

       [ -f /var/lock/subsys/crond ] && restart || :

       ;;

  *)

       echo $"Usage: $0 {start|stop|status|reload|restart|condrestart}"

       exit 1

esac

[root@web init.d]#