| SYSTEM INFORMATION | |
|---|---|
| OS type and version | Debian 13 |
| Webmin version | 8.1.0 |
MariaDB version 11.8.6
MariaDB System Variables:
**Fatal Error!**
/usr/libexec/mariadb
in MariaDB Server Configuration:
**Fatal Error!**
Could not find [mysqld] section in my.cnf
The server is empty, just installed. Is that normal or I have to do something?
That’s not normal.
What appears in /var/log/miniserv.error when you visit this page and get this error?
Mine is in /var/webmin
Oops. Sorry. Yeah, it’s /var/webmin/miniserv.error.
It should be in /var/log/webmin, but we have to write some code for that to happen.
Joe…, is here what I found:
[26/Mar/2026:19:57:29 +0000] Reloading configuration
[26/Mar/2026:19:57:44 +0000] [my ip] /mysql/list_vars.cgi : /usr/libexec/mariadb
[26/Mar/2026:19:57:51 +0000] [my ip] /mysql/edit_cnf.cgi : Could not find [mysqld] section in my.cnf
OK, so that’s strange.
This was a freshly installed system? Nothing was preinstalled?
What’s in /etc/my.cnf and what’s in /etc/my.cnf.d/mariadb-server.cnf (if it exists, I don’t know what the included file is called on Debian…but, I assume there is a server cnf file in /etc/my.cnf.d of some sort. I’d like to see that.
Joe, is fresh but not, I setup yesterday .., save a snapshot.. today I deploy in a different IP and Virtualmin tell me that, fix the IP. Everything ok.
I almost forgot, I used that “Re-setup” feature… I forgot the name. And everything asked me again about RAM, etc.
Right now Im rebuild from zero.. using:
sudo sh -c "$(curl -fsSL https://download.virtualmin.com/virtualmin-install)" -- --bundle LAMP
Just for the record on my Debian 13 machine it has an empty entry. Just the [mysql]:
root@never:/etc/mysql/conf.d# cat mysql.cnf
[mysql]You don’t have an /etc/my.cnf?
Yeah. Lots of comment. The only active section.
[client-server]
# Port or socket location where to connect
# port = 3306
socket = /run/mysqld/mysqld.sock
# Import all .cnf files from configuration directory
!includedir /etc/mysql/conf.d/
!includedir /etc/mysql/mariadb.conf.d/
~Joe, just reinstall and the error persist:
Fatal Error!
/usr/libexec/mariadb
In: MariaDB System Variables
Regards
Kazuma
Are you renting space for this? If so, are they providing the image? It could be the provider is doing something?
You need to look at the included files.
I didn’t tell you to reinstall. Makes no sense to do the same thing over and over and expect different results.
Testing on Vultr, first off I see issues during install with firewalld. (is this a known issue as I don’t use Debian)
I also see the error at variables section
I reinstalled it just to make sure it wasn’t some problem I caused during the first installation. Now I’m certain the problem is a bug in the software or an issue installing on Debian. Either way, it was enough to make me take a step back before considering migrating.
Would Debian 12 be better without this bug?
See what joe has to say, they might be a quick patch.
It would be better to use 13 to save upgrades later on.
I just reinstall using Debian12 and no error on that MariaDB.
I usually don’t perform updates; I’ve been in this field since 1995 and learned through hard work. I prefer not to update or allow updates to occur. I set up the server, test it, test it again, and if it’s okay, it stays that way. Security?! I prefer to use a good external firewall, preferably hardware-based, limit port forwarding by applications, and we live without security updates. Nowadays, especially with AI involved, the best thing is to have a well-isolated server, and even then we know it will never be completely secure. So, between partially secure and stable with less security, I prefer the second option. So whether it’s Debian 12 or 13, as long as it works for me, it’s okay. Regarding Joe, he didn’t ask me to wait for a patch; he scolded me for reinstalling without his permission. Lol!
