Centos and Red Hat notes, part 1

Red Hat logo

Part 1
Part 2
Part 3

I’ve decided to take the exams for Red Hat Certified Systems Administrator (RHCSA) and Red Hat Certified Engineer (RHCE). I’ve been meaning to do this for 10 or 15 years! This post is just notes for myself.

I usually use Ubuntu so these notes are specifically for learning CentOS.

Users, passwords, etc.

  • list users. This can be done with more /etc/passwd. The output of this includes username, password, UID, GID, User Details (eg first name), home directory, and default shell. You could also use getent passwd since getent is a tool that searches databases on the system, like hosts and services, and others.
  • create user. This will create a user named john with a uid/gid of 2000
    • sudo useradd -u 2000 john
    • sudo passwd john #enter password for john here
  • set password validity for user. The chage command fordes users to change passwords to comply with password-aging policies
  • change password. passwd
  • change root password. sudo passwd root
  • change root password when you don’t have sudo privilege. You need physical access to machine.
    • reboot machine and press e during GRUB loader screen. That will open an editor with current kernel boot options.
    • find line starting with linux16 and add rd.break at the end and hit Ctrl+X to exit
      mount -o remount,rw /sysroot/
      chroot /sysroot
      passwd to change your password
      touch /.autorelabel
      exit
      exit or reboot
  • sudoers file. Your user must be in this if you want to run sudo in front of your commands.
su - root                       #switch to the root user
#Option 1: add the user to the 'wheel' group
usermod -a -G wheel centos      #add the user 'centos' to the group 'wheel' which has the rights to run the sudo command
#Option 2: add the user directly to /etc/sudoers file
visudo
#add a line like this to the file 'centos ALL=(ALL:ALL) ALL'
  • change hostname
    • nmtui or
    • edit /etc/hostname or
    • use the command hostnamectl set-hostname SOME_NAME or
    • use nmcli: nmcli general hostname SOME_NAME
      NB: if you want to change the ‘pretty hostname’ without restarting the OS, you can use hostnamectl set-hostname SOME_NAME --pretty or create/edit the file /etc/machine-info so that it has the pretty hostname in it.

Networking, routing, etc

  • My Hyper-V on Win10. When I deployed Centos 7.9.2009 on Hyper-V using the ISO found here, I typed nmcli and it showed the interface eth0 was in a disconnected state. Using the console UI in Hyper-V, I typed nmcli device connect eth0 and then the interface obtained a DHCP address immediately.

  • Set a static IP. Edit /etc/sysconfig/network-scripts/ifcfg-eth0. I’m experienced enough not to type everything out, but in my case I’ll note that BOOTPROTO can be changed from dhcp to static and that ONBOOT can be changed to yes. My file now looks like this:

TYPE=Ethernet
PROXY_METHOD=none
BROWSER_ONLY=no
BOOTPROTO=static        #edited from 'dhcp' to 'static'
DEFROUTE=yes
IPV4_FAILURE_FATAL=no
IPV6INIT=yes
IPV6_AUTOCONF=yes
IPV6_DEFROUTE=yes
IPV6_FAILURE_FATAL=no
IPV6_ADDR_GEN_MODE=stable-privacy
NAME=eth0
UUID=3abc8a88-85fe-4a80-8e29-ed1ff1dcf739
DEVICE=eth0
ONBOOT=yes              #edited from 'no' to 'yes'
IPADDR=172.23.110.145   #I added this line
NETMASK=255.255.240.0   #I added this line
GATEWAY=172.23.96.1     #I added this line
DNS1=8.8.8.8            #I added this line
DNS2=8.8.4.4            #I added this line
  • On Centos, use the ip route command (not just route like Ubuntu)
  • Add a secondary IP address.
    Option 1: nmtui and follow the interface to add a secondary IP. Run service network restart after that.

    Option 2: Edit the config files directly. Can only be done if NM_CONTROLLED="no" or is not configured at all on the interface. You would just add these lines to the above config file:
IPADDR1=172.23.110.146
PREFIX1=20
NETMASK1=255.255.240.0
  • Add a temporary secondary IP address. This will last only until server reboot our the next network service restart.
ip a add 172.23.110.146/20 dev eth0
  • restart network service: service network restart

SE Linux

  • install SE Linux. This worked for me on Centos 7.9
sudo yum install policycoreutils policycoreutils-python setools setools-console setroubleshoot
  • disabled vs enforcing
    The file /etc/selinux/config should have SELINUX=enforcing or permissive or disabled (editing this file and rebooting is the way to ensure that the mode will persist after reboot)
  • getenforce - this command will tell you which mode you are in
  • sudo setenforce 0 for permissive or sudo setenforce 1 for enforcing (does not persist after reboot)
  • sestatus to check status of SE Linux

Apache

  • sudo yum install -y httpd - this will install Apache
  • remember to allow service to interact with network through firewall
#notice the '--permament' option (in order to save rule to survive during reboots)
firewall-cmd --permanent --add-service=http
firewall-cmd --reload
  • remember to start a service after installing it, and to enable it (to autostart after reboot)
systemctl enable httpd
systemctl start httpd

Updated:

You May Also Enjoy