Job for mariadb.service failed bug fixed

Last month I set up a development server using Fedora 20. On that server I wanted to change MySQL socket file from /var/lib/mysql/mysql.sock to /dev/shm/mysql.sock. Basically I wanted to move MySQL socket file to shared memory for performance gain.

I made the change in /etc/my.cnf, but when I tried to start mysqld, I got the error message:

Job for mariadb.service failed. See ‘systemctl status mariadb.service’ and ‘journalctl -xn’ for details.

That was my first time heard about mariadb. I did some background research on mariadb and the relationship between mysqld and mariadb. But that did not help to solve the problem. So I had to revert to the default MySQL socket file /var/lib/mysql/mysql.sock.

Today I had some free minutes and recalled this unresolved issue. To my surprise, I just changed

socket=/var/lib/mysql/mysql.sock

to

socket=/dev/shm/mysql.sock

in /etc/my.cnf. And this time mysqld can start without any errors.

What changes have I made to the system? Nothing but a yum update.

So I assume there was a bug in mariadb (or mysqld) which is fixed by the recent release.

Mariadb updated
Mariadb updated

Leave a Reply

Your email address will not be published. Required fields are marked *