Installing MySQL 9.7 LTS Community Edition on CentOS

Installing MySQL 9.7 LTS Community Edition on CentOS

Historically installing MySQL on a RedHat Compatible Linux server was as simple as yum install mysql-server.

Today’s MySQL Oracle Linux, Red Hat Enterprise Linux, CentOS, and Fedora 9.7 instructions are not accurate mixing in 8.4 and 8.0 information, and GenAI can also mislead you for the installation process. The most notable issue that I’d like to resolve is correct GPG Key checking, currently this is disabled --nogpgcheck as 2025 key does not operate as expected.

The following is my cloud-init.yml file I use to build VMs with OrbStack . In summary:

  • Define correct yum repos
  • Install with GPG disabled
  • Enable Service
  • Set a usable (aka non-temporary) root password

cloud-init.yml

#cloud-config
# MySQL 9.7 LTS install for CentOS Stream 9 (works on aarch64/ARM64).
#
# Notes:
#  - The repo URL path is mysql-9.7-community (NOT mysql-9.7-lts-community);
#    only the repo *id* carries the -lts- label.
#  - MySQL's published GPG key file still shows expired (2025-10-22) even
#    though the key was re-extended, and the RPM-GPG-KEY-mysql-2025 URL is
#    empty. So gpgcheck is disabled here to keep an unattended boot from
#    failing. Re-enable once MySQL refreshes the published key file.

write_files:
  - path: /etc/yum.repos.d/mysql-community.repo
    permissions: '0644'
    owner: root:root
    content: |
      [mysql-9.7-lts-community]
      name=MySQL 9.7 LTS Community Server
      baseurl=https://repo.mysql.com/yum/mysql-9.7-community/el/9/$basearch/
      enabled=1
      gpgcheck=0

      [mysql-connectors-community]
      name=MySQL Connectors Community
      baseurl=https://repo.mysql.com/yum/mysql-connectors-community/el/9/$basearch/
      enabled=1
      gpgcheck=0

      [mysql-tools-community]
      name=MySQL Tools Community
      baseurl=https://repo.mysql.com/yum/mysql-tools-community/el/9/$basearch/
      enabled=1
      gpgcheck=0

runcmd:
  - dnf clean all
  - dnf install -y --nogpgcheck mysql-community-server
  - systemctl enable --now mysqld
  - mysql --version
  # Rotate the generated temporary root password to a fresh random one.
  - [ sh, -c, "OLD_PASSWD=$(grep 'temporary password' /var/log/mysqld.log | awk '{print $NF}'); PASSWD=\"#A$(date | md5sum | cut -c1-12)\"; echo \"ALTER USER root@localhost IDENTIFIED BY '${PASSWD}';\" | mysql -uroot -p\"${OLD_PASSWD}\" --connect-expired-password; echo \"${PASSWD}\" > /root/.mysql.password; chmod 600 /root/.mysql.password" ]

Installation Demo

Full installation instructions with orb installed and using mysql97.yml

wget https://gist.githubusercontent.com/ronaldbradford/672ebd49fc0ab0a092f74bec23461ff6/raw/9424a0f8b54b5b79356182aa8f119676cf60280f/mysql97.yml
VM_NAME="centos10"
orb create centos ${VM_NAME} -c mysql97.yml
orb -m ${VM_NAME} tail /var/log/cloud-init-output.log
ssh ${VM_NAME}@orb
PASSWD=$(sudo cat /root/.mysql.password)
mysql -uroot -p${PASSWD} -e "SELECT VERSION()"
Tagged with: MySQL 9.7 Installation

Producing One-Sample Z-Test statistics with SQL

The one-sample Z-test determines whether a sample mean differs significantly from a known population mean when the population standard deviation is also known. It is the appropriate test when the population parameters are established — quality control benchmarks, national averages, long-run process measurements — and you want to evaluate whether a new sample is consistent with them.

Using GenAI directly in the database. A practical example using MySQL 8.0

If you have a typical MySQL production setup using MySQL 8.0 (EOL) with replication, you can take advantage of VillageSQL extensions to generate AI responses directly with your source data with no impact on your production setup or existing application software.

Producing Skewness statistics with SQL

Skewness measures the asymmetry of a distribution. A perfectly symmetric distribution has a skewness of zero. A positive skew (right-skewed) means the tail extends to the right — a small number of high values pull the mean above the median.