MariaDB service fails after Leap16 installation

Hi!
I have a headless home server that was running fine with openSUSE 15.6. I recently upgraded it to openSUSE 16 and since then mariadb.service fails to start.
systemctl status mariadb reads:

Process: 1711 ExecStartPre=/usr/libexec/mysql/mysql-systemd-helper install (code=exited, status=0/SUCCESS)
Process: 1754 ExecStartPre=/usr/libexec/mysql/mysql-systemd-helper upgrade (code=exited, status=0/SUCCESS)
Process: 1767 ExecStart=/usr/libexec/mysql/mysql-systemd-helper start (code=exited, status=1/FAILURE)

If I try to start the service manually (sudo systemctl start mariadb.service) it also fails, and advises me to retrieve more information from journalctl -xeu mariadb.service, which I did; what seems relevant is:

jan 25 13:50:29 domus mysql-systemd-helper[4120]: Could not open required defaults file: /etc/my.cnf
jan 25 13:50:29 domus mysql-systemd-helper[4120]: Fatal error in defaults handling. Program aborted
jan 25 13:50:29 domus systemd[1]: mariadb.service: Main process exited, code=exited, status=1/FAILURE

However, the file is there and seems to have the right owner and attributes. If I type ls -l /etc/my.cnf, I get:

-rw-r--r--. 1 mysql mysql 3455 jan 24 21:37 /etc/my.cnf

More astonishing to me is, if I manually force the failing command to start, using runuser -u mysql /usr/libexec/mysql/mysql-systemd-helper start, it starts without visible failures and MariaDB server answers my SQL requests.

Of course, the goal is to start MariaDB automatically during startup, and not having to log in to server and manually start MariaDB every time I need to access my data.

I am a bit lost here, and I would appreciate hints.

A recent german thread:

Probably SELinux related:
https://bugzilla.opensuse.org/show_bug.cgi?id=1255024

Thanks for the hint!
I read both posts (fortunately I am able to speak German), but I couldn’t yet find a solution. I noticed however, that:

sudo setenforce 0

let me start the service. Unfortunately, this command doesn’t survive a reboot.

I suppose I must get a crash course in SELinux…

Have you tried the solutions posted in the bugreport?

I must confess, in the bug report I could only see one hint: restorecon -Rv {/var/,}/run/mysql/; and no, that didn’t help.

https://bugzilla.opensuse.org/show_bug.cgi?id=1255024#c17