ssh peter@stdb01
# check status of mariadb, mine showed:
[peter@stdb01 ~]$ systemctl status mariadb
â—‹ mariadb.service - MariaDB 10.5 database server
Loaded: loaded (/usr/lib/systemd/system/mariadb.service; enabled; preset: disabled)
Active: inactive (dead) since Tue 2025-09-02 07:24:49 UTC; 3min 43s
ago
[peter@stdb01 ~]$ systemctl start mariadb
Failed to start mariadb.service: Access denied
See system logs and 'systemctl status mariadb.service' for details.
I checked journalctl -xeu for logs. turns out my /var/run/mariadb folder had some wrong permissions.
I ran:
sudo chown -R mysql:mysql /var/run/mariadb
sudo chmod 755 /var/run/mariadb
sudo systemctl restart mariadb
It looks like MariaDB is starting up and then immediately shutting down, which usually points to either a config problem or corruption in the data files. From your screenshot the service itself loads fine, but the main process exits right away with “MariaDB server is down.” First thing I’d do is check the detailed error logs with journalctl -xeu mariadb and also look under /var/lib/mysql/*.err for any startup errors that aren’t showing in systemctl.
A few common causes here are bad or conflicting settings in /etc/my.cnf, something else already binding to port 3306, or corruption in the InnoDB tablespace/redo logs that forces the server to quit. If it’s config related you can run mysqld --help --verbose to validate options. If it looks like corruption, try adding innodb_force_recovery=1 under [mysqld] in the config to see if it will start, and increment that value if needed.
If the database does start with forced recovery you’ll want to dump your data right away, since that mode isn’t for production use. And if you find that the ibdata or .ibd files are actually corrupted beyond what InnoDB can recover, a specialized repair tool like Stellar Repair for MySQL can be handy for salvaging and rebuilding the affected tables so you don’t lose data.