Linux shell脚本，按顺序批量启动多个jar包，批量启动spring cloud的jar包
一. 手动一个一个启动的方式：
nohup java -jar eurekaserver.jar > ../logs/eurekaserver.log 2>&1 & nohup java -jar configserver.jar > ../logs/configserver.log 2>&1 & nohup java -jar zuulgatewayserver.jar > ../logs/zuulgatewayserver.log 2>&1 & nohup java -jar authorizeserver.jar > ../logs/authorizeserver.log 2>&1 & nohup java -jar logstatisticsserver.jar > ../logs/logstatisticsserver.log 2>&1 & nohup java -jar appmgrserver.jar > ../logs/appmgrserver.log 2>&1 & nohup java -jar organizeuserserver.jar > ../logs/organizeuserserver.log 2>&1 & nohup java -jar jtmopserver.jar > ../logs/jtmopserver.log 2>&1 &
二. shell脚本代码：
#!/bin/sh export EUREKA=eurekaserver.jar export CONFIG=configserver.jar export GATEWAY=zuulgatewayserver.jar export AUTH=authorizeserver.jar export LOG=logstatisticsserver.jar export APPMGR=appmgrserver.jar export ORGANIZEUSER=organizeuserserver.jar export JTMOP=jtmopserver.jar export EUREKA_log=../logs/eurekaserver.log export CONFIG_log=../logs/configserver.log export GATEWAY_log=../logs/zuulgatewayserver.log export AUTH_log=../logs/authorizeserver.log export LOG_log=../logs/logstatisticsserver.log export APPMGR_log=../logs/appmgrserver.log export ORGANIZEUSER_log=../logs/organizeuserserver.log export JTMOP_log=../logs/jtmopserver.log export EUREKA_port=10080 export CONFIG_port=10085 export GATEWAY_port=10084 export AUTH_port=10088 export LOG_port=10093 export APPMGR_port=10092 export ORGANIZEUSER_port=10087 export JTMOP_port=10091 case "$1" in start) ## 启动eureka echo "--------eureka 开始启动--------------" nohup java -jar $EUREKA > $EUREKA_log 2>&1 & EUREKA_pid=`lsof -i:$EUREKA_port|grep "LISTEN"|awk '{print $2}'` until [ -n "$EUREKA_pid" ] do EUREKA_pid=`lsof -i:$EUREKA_port|grep "LISTEN"|awk '{print $2}'` done echo "EUREKA pid is $EUREKA_pid" sleep 10 echo "--------eureka 启动成功--------------" ## 启动config echo "--------开始启动CONFIG---------------" nohup java -jar $CONFIG > $CONFIG_log 2>&1 & CONFIG_pid=`lsof -i:$CONFIG_port|grep "LISTEN"|awk '{print $2}'` until [ -n "$CONFIG_pid" ] do CONFIG_pid=`lsof -i:$CONFIG_port|grep "LISTEN"|awk '{print $2}'` done echo "CONFIG pid is $CONFIG_pid" sleep 30 echo "---------CONFIG 启动成功-----------" ## 启动gateway echo "--------开始启动GATEWAY---------------" nohup java -jar $GATEWAY > $GATEWAY_log 2>&1 & GATEWAY_pid=`lsof -i:$GATEWAY_port|grep "LISTEN"|awk '{print $2}'` until [ -n "$GATEWAY_pid" ] do GATEWAY_pid=`lsof -i:$GATEWAY_port|grep "LISTEN"|awk '{print $2}'` done echo "GATEWAY pid is $GATEWAY_pid" echo "---------GATEWAY 启动成功-----------" ## 启动auth echo "--------开始启动AUTH---------------" nohup java -jar $AUTH > $AUTH_log 2>&1 & AUTH_pid=`lsof -i:$AUTH_port|grep "LISTEN"|awk '{print $2}'` until [ -n "$AUTH_pid" ] do AUTH_pid=`lsof -i:$AUTH_port|grep "LISTEN"|awk '{print $2}'` done echo "AUTH pid is $AUTH_pid" echo "---------AUTH 启动成功-----------" ## 启动log echo "--------开始启动LOG---------------" nohup java -jar $LOG > $LOG_log 2>&1 & LOG_pid=`lsof -i:$LOG_port|grep "LISTEN"|awk '{print $2}'` until [ -n "$LOG_pid" ] do LOG_pid=`lsof -i:$LOG_port|grep "LISTEN"|awk '{print $2}'` done echo "LOG pid is $LOG_pid" echo "---------LOG 启动成功-----------" ## 启动appmgr echo "--------开始启动APPMGR---------------" nohup java -jar $APPMGR > $APPMGR_log 2>&1 & APPMGR_pid=`lsof -i:$APPMGR_port|grep "LISTEN"|awk '{print $2}'` until [ -n "$APPMGR_pid" ] do APPMGR_pid=`lsof -i:$APPMGR_port|grep "LISTEN"|awk '{print $2}'` done echo "APPMGR pid is $APPMGR_pid" echo "---------APPMGR 启动成功-----------" ## 启动organizeuser echo "--------开始启动ORGANIZEUSER---------------" nohup java -jar $ORGANIZEUSER > $ORGANIZEUSER_log 2>&1 & ORGANIZEUSER_pid=`lsof -i:$ORGANIZEUSER_port|grep "LISTEN"|awk '{print $2}'` until [ -n "$ORGANIZEUSER_pid" ] do ORGANIZEUSER_pid=`lsof -i:$ORGANIZEUSER_port|grep "LISTEN"|awk '{print $2}'` done echo "ORGANIZEUSER pid is $ORGANIZEUSER_pid" echo "---------ORGANIZEUSER 启动成功-----------" ## 启动jtmop echo "--------开始启动JTMOP---------------" nohup java -jar $JTMOP > $JTMOP_log 2>&1 & JTMOP_pid=`lsof -i:$JTMOP_port|grep "LISTEN"|awk '{print $2}'` until [ -n "$JTMOP_pid" ] do JTMOP_pid=`lsof -i:$JTMOP_port|grep "LISTEN"|awk '{print $2}'` done echo "JTMOP pid is $JTMOP_pid" echo "---------JTMOP 启动成功-----------" echo "===startAll success===" ;; stop) P_ID=`ps -ef | grep -w $EUREKA | grep -v "grep" | awk '{print $2}'` if [ "$P_ID" == "" ]; then echo "===EUREKA process not exists or stop success" else kill -9 $P_ID echo "EUREKA killed success" fi P_ID=`ps -ef | grep -w $CONFIG | grep -v "grep" | awk '{print $2}'` if [ "$P_ID" == "" ]; then echo "===CONFIG process not exists or stop success" else kill -9 $P_ID echo "CONFIG killed success" fi P_ID=`ps -ef | grep -w $GATEWAY | grep -v "grep" | awk '{print $2}'` if [ "$P_ID" == "" ]; then echo "===GATEWAY process not exists or stop success" else kill -9 $P_ID echo "GATEWAY killed success" fi P_ID=`ps -ef | grep -w $AUTH | grep -v "grep" | awk '{print $2}'` if [ "$P_ID" == "" ]; then echo "===AUTH process not exists or stop success" else kill -9 $P_ID echo "AUTH killed success" fi P_ID=`ps -ef | grep -w $LOG | grep -v "grep" | awk '{print $2}'` if [ "$P_ID" == "" ]; then echo "===LOG process not exists or stop success" else kill -9 $P_ID echo "LOG killed success" fi P_ID=`ps -ef | grep -w $APPMGR | grep -v "grep" | awk '{print $2}'` if [ "$P_ID" == "" ]; then echo "===APPMGR process not exists or stop success" else kill -9 $P_ID echo "APPMGR killed success" fi P_ID=`ps -ef | grep -w $ORGANIZEUSER | grep -v "grep" | awk '{print $2}'` if [ "$P_ID" == "" ]; then echo "===ORGANIZEUSER process not exists or stop success" else kill -9 $P_ID echo "ORGANIZEUSER killed success" fi P_ID=`ps -ef | grep -w $JTMOP | grep -v "grep" | awk '{print $2}'` if [ "$P_ID" == "" ]; then echo "===JTMOP process not exists or stop success" else kill -9 $P_ID echo "JTMOP killed success" fi echo "===stop success===" ;; restart) $0 stop sleep 10 $0 start echo "===restart success===" ;; esac exit 0
三. shell脚本批量启动、批量停止、批量重启等步骤：
1. 脚本名：startupAll.sh
2. 先赋予脚本可执行权限：
chmod +x startupAll.sh
3. 执行批量启动的shell脚本命令：
./startupAll.sh start
4. 执行批量停止的shell脚本命令：
./startupAll.sh stop
5. 执行批量重启的shell脚本命令：
./startupAll.sh restart
四. 可能出现如下报错：
-bash: ./xxx.sh: /bin/sh^M: bad interpreter: No such file or directory
错误介绍：
通过上面我们可以看到有个^M个标志，这个标志是：Windows环境下dos格式文件传输到unix系统时,会在每行的结尾多一个^M，所以在执行的时候出现了这种现象，但是你在unix或者Linux环境下使用vi编辑的时候，会在下面显示此文件的格式，比如” startupAll.sh” [dos] 2L,20C字样，表示这是一个【dos】的格式文件，如果是MAC的系统则会出现【MAC】的字样，因为文件格式的原因，有时候我们是unix程序或shell程序，则就要把dos文件转化为unix的文件格式。
错误解决方式：有如下四种解决方式
解决方式1：
也是最简单的方法,直接在当前的文件目录下执行下面命令
sed -i -e 's/ $//' startupAll.sh
解决方式2：
使用unix/Linux中的vi命令，更改fileformat格式
首先执行：
vi startupAll.sh
使用vi startupAll.sh查看时在最后一行会出现了下面的
"startupAll.sh" [dos] 2L, 19C
说明这个文件编码方式是windows编辑的，必须转化格式为unix格式 ,按ESC键，再次执行下面命令后
:set ff
敲回车，出现如下
fileformat=dos
修改为下面
vi test.sh :set ff=unix :x
解决方式3：
使用dos2unix和unix2dos命令
Linux提供了两种文本格式相互转化的命令：dos2unix和unix2dos，dos2unix把” ”转化成” ”，unixtodos把” ”转化成” ”。Java程序中，使用 Runtime.getRuntime().exec(“unix2dos “+ fileName);来实现对文本格式互转命令的调用。多说一句，执行指令前要先看看linux上是否有安装unix2dos/dos2unix工具。
1.首先查看是否安装了这个工具，使用grep正则使用
yum list installed | grep -nE 'dos2unix|dos2'
如果安装了这个工具，则执行下面命令
dos2unix startupAll.sh
出现下面显示说明转化ok
[root@hostname webapps]# dos2unix j.sh
dos2unix: converting file j.sh to Unix format ...
[root@hostname webapps]#
解决方式4：
在编辑代码的时候设置编辑器的line endings编码方式
如果你在Windows或Mac上使用sublime文本编辑脚本:单击View > Line ending > Unix并再次保存文件。例如：
在notepad++中，您可以通过按下按钮为文件设置它 Edit –> EOL Conversion –> UNIX/OSX Format
Eclipse用户,您可以更改文件编码直接从 File > Convert Line Delimiters To > Unix (LF, , 0Α, ¶)
或者更改 Unix on Window > Preferences > General > Workspace panel:
通过以上4中方式，选择其中一种即可，所以以后要注意这个文件的格式最好把你使用的编辑器设置为unix格式。