linux 自动重启服务 shell

linux下监视进程崩溃挂掉后自动重启的shell脚本 - Python技术站 (pythonjishu.com)

新建日志存放目录：

 mkdir /mnt/data/logs/monitor -p

脚本路径：

/mnt/data/zzy/scripts

案例一：监控nginx

1:脚本：cat monitor_nginx.sh

#!/bin/bash
PROCESS_NAME="nginx"
使用pgrep命令来查找进程id
PROCESS_ID=$(pgrep "${PROCESS_NAME}")
if [ "${PROCESS_ID}" = "" ]; then
echo  $(date +%T%n%F) "Nginx is not running. Starting it..."
#/etc/init.d/nginx start
systemctl start nginx
else
echo    "Nginx is running. Checking if it's healthy..."
# 检测进程是否正常运行
RESULT=$(kill -0 "${PROCESS_ID}" 2>&1)
if [[ "${RESULT}" =~ "No such process" ]]; then
echo $(date +%T%n%F) "Nginx is not running. Restarting it..."
# 结束进程
kill -9 "${PROCESS_ID}"
# 启动进程
/etc/init.d/nginx start
systemctl start nginx
else
echo "Nginx is running normally."
fi
fi

2- 定时任务：

vim /etc/crontab

5分钟检测一次

*/5 * * * * root sh /mnt/data/zzy/scripts/monitor_nginx.sh  >>/mnt/data/logs/monitor/nginxstatus.log 2>&1

cat /mnt/data/logs/monitor/nginxlogfile.log

案例二：监控mysql

1- 脚本 cat monitor_mysql.sh

#!/bin/bash
PROCESS_NAME="mysqld"
使用pgrep命令来查找进程id
PROCESS_ID=$(pgrep "${PROCESS_NAME}")
if [ "${PROCESS_ID}" = "" ]; then
echo  $(date +%T%n%F) "MySQL is not running. Starting it..."
#/etc/init.d/mysql start
systemctl start mysql
else
echo   "MySQL is running. Checking if it's healthy..."
# 检测进程是否正常运行
RESULT=$(kill -0 "${PROCESS_ID}" 2>&1)
if [[ "${RESULT}" =~ "No such process" ]]; then
echo $(date +%T%n%F) "MySQL is not running. Restarting it..."
# 结束进程
kill -9 "${PROCESS_ID}"
# 启动进程
# /etc/init.d/mysql start
systemctl start mysql
else
echo "MySQL is running normally."
fi
fi

2- 定时任务：

vim /etc/crontab

*/5 * * * * root sh  /mnt/data/zzy/scripts/monitor_mysql.sh >>/mnt/data/logs/monitor/mysqlstatus.log 2>&1

案例三：监控docker

1-脚本

cat monitor_docker.sh

#!/bin/bash
PROCESS_NAME="docker"
使用pgrep命令来查找进程id
PROCESS_ID=$(pgrep "${PROCESS_NAME}")
if [ "${PROCESS_ID}" = "" ]; then
echo  $(date +%T%n%F) "Docker is not running. Starting it..."
#/etc/init.d/mysql start
systemctl start docker
else
echo   "Docker is running. Checking if it's healthy..."
# 检测进程是否正常运行
RESULT=$(kill -0 "${PROCESS_ID}" 2>&1)
if [[ "${RESULT}" =~ "No such process" ]]; then
echo $(date +%T%n%F) "Docker is not running. Restarting it..."
# 结束进程
kill -9 "${PROCESS_ID}"
# 启动进程
# /etc/init.d/mysql start
systemctl start docker
else
echo "Docker  is running normally."
fi
fi

2- 定时任务

vim /etc/crontab

*/5 * * * * root sh  /mnt/data/zzy/scripts/monitor_docker.sh >>/mnt/data/logs/monitor/dockerstatus.log 2>&1

案例四： nacos（不太一样）

cat monitor_nacos.sh

#!/bin/bash
PROCESS_NAME="nacos"
使用pgrep命令来查找进程id
#PROCESS_ID=$(pgrep "${PROCESS_NAME}")
PROCESS_ID=$(ps -ef | grep nacos.nacos | grep -v grep | awk '{print $2}')
if [ "${PROCESS_ID}" = "" ]; then
echo  $(date +%T%n%F) " $PROCESS_NAME is not running. Starting it..."
#/etc/init.d/nginx start
/mnt/data/nacos/nacos/bin/startup.sh -m standalone
else
echo    "$PROCESS_NAME is running. Checking if it's healthy..."
# 检测进程是否正常运行
RESULT=$(kill -0 "${PROCESS_ID}" 2>&1)
if [[ "${RESULT}" =~ "No such process" ]]; then
echo $(date +%T%n%F) "$PROCESS_NAME is not running. Restarting it..."
# 结束进程
kill -9 "${PROCESS_ID}"
# 启动进程
/mnt/data/nacos/nacos/bin/startup.sh -m standalone
else
echo "$PROCESS_NAME is running normally."
fi
fi