Hi,

Today I have updated a CentOS 6.8 system that has MariaDB-server installed from
http://yum.mariadb.org/5.5/centos6-amd64 and found that I had an SELinux issue when I tried to restart the service.

Using the information at https://wiki.centos.org/HowTos/SELinux I have created a local policy that seems to fix things

module marialocal 1.0;

require {
        type mysqld_safe_t;
        class capability { setuid setgid };
}

#============= mysqld_safe_t ==============

allow mysqld_safe_t self:capability setgid;
allow mysqld_safe_t self:capability setuid;

This seems to fix things for me, but I thought I had better see if others are experiencing the same problem ?

It may be useful to know that the particular rpms are MariaDB-server-5.5.54-1.el6.x86_64 and
selinux-policy-3.7.19-292.el6_8.2.noarch.

If it is a bigger issue than just myself, should I report this somewhere else to see if we can get a fix added to the next MariaDB-server rpm ?

Thanks