Solutions for monitoring disk usage

Description: There are 100 centos servers that communicate with each other via intranet ssh, the user name and password are known, and the server ip list is known. Please design a technical solution for the server with an output disk usage rate exceeding 90%. It is required to write a detailed technical solution, please write the code for the commands and scripts involved, and please introduce the installation instructions for the third-party libraries.

Program

Program ideas:

• Test if the mailbox is available

• Set a scheduled task to execute the following script every five minutes to view the occupancy

• Send alarm email when it is found that the occupancy ratio exceeds 90%

• Install ansible service, compile yml file to execute tasks in batch

Test mail

#Here we use qq mailbox as the recipient
vim /etc/mail.rc
set bsdcompat
set from = [email protected] #receiver name
set smtp=smtp.qq.com  #smtp地址
set smtp -auth-user = [email protected] #recipient address  
set smtp -auth-password = xxxxxxx #smtp verification code, operate in the qq mailbox, send the information through the mobile phone to obtain the verification code
set smtp-auth=login


#test
echo  " Mail Test (Test Content) " | mail -s " Test Results (Subject) " [email protected]

Script content

vim /server/scripts/df.sh
#!/bin/bash                                                                             
>/tmp/disk.log
#Here is the test threshold
D_Value=90
#Remove disk percentage
D_Use=`df -h|grep '^/dev'|awk '{print $5}'|sed 's#%##'`
#Remove the host private IP address
IP_Addr=`/sbin/ifconfig eth1|grep inet|grep -v 127.0.0.1|grep -v inet6|awk '{print $2}'|tr -d "addr:"`
for i in $D_Use
do
    if [  $i -ge $D_Value ];then
        D_Name=`df -h|sed -n '/'"$i"'/p'|awk '{print $1}'`
        D_Part = ` df -h | sed -n ' / '" $ i "' / p ' | awk  ' {print $ 6} ' `
         date >> / tmp / disk.log
         echo  " $ IP_Addr server's $ D_Name (partition $ D_Part) The usage rate exceeds $ D_Value%, and the current usage rate is $ i%. " >> / tmp / disk.log
     fi 
done 
if [-s /tmp/disk.log]; then 
    cat /tmp/disk.log|mail -s " Warning-Used DISK "  1354586675 @ qq.com
 fi

Scheduled tasks

crontab -e
#Execute the script every five minutes
*/5 * * * * /server/scripts/df.sh

Deploy ansible

yum  install warm-release - and
 yum  install ansible - y
ansible --version

#Set the host list, use sequence matching method
vim /etc/ansible/hosts
[ubuntu_server]
172.16.1.[41:100]


#Distribute keys to 100 hosts
ssh-keygen
ansible ubantu_server -m copy -a "src=/root .ssh/id_rsa.pub dest=/root.ssh"


mkdir  /etc/ansible/palybook

#Write yml file
vim df .yml
 - hosts: ubantu_server
  remote_user: root
  tasks:
  - name: Create a script directory
     File :
      path: /server/scripts
      state: directory
      mode: 0755  
  - name: push script
    copy:
      src: / server / scripts / df . sh 
      dest: / server / scripts /
  - name: Push mail profile
    copy:
      src: /etc/mail.rc
      dest: / etc /
      force: yes
  - name: create a scheduled task
    cron:
      name: "Disk percentage check"
      minute: "*/5"
      job: "sh /server/scripts/df.sh"
      
      
  #Execution script
  ansible -playbook ubantu.yml