MySQL - All of a sudden "Communications link failure" between docker container and MySQL

L. Don :

I'm working on a Docker application that uses MySQL to store very large databases (that's due to legacy reasons). This one is installed on the host machine.

Today I was doing common jobs that I've been doing for 1 or 2 months, and all of a sudden I can't communicate anymore with my database.

The uri of the db has been the same for months: jdbc:mysql://localhost:3306/dbname?verifyServerCertificate=false&useSSL=true&useJDBCCompliantTimezoneShift=true&useLegacyDatetimeCode=false&serverTimezone=UTC

As soon as i try to connect to the database from my application that runs on the Virtual Machine I receive:

java.sql.SQLException: Cannot create PoolableConnectionFactory (Communications link failure
The last packet sent successfully to the server was 0 milliseconds ago. The driver has not received any packets from the server.)
...
Caused by: java.net.ConnectException: Connection timed out

The guilty code is this:

private static boolean checkForExistence(String dbName) {
    boolean exists = false;
    BasicDataSource dataSource = new BasicDataSource();

    dataSource.setDriverClassName(JDBC_DRIVER); //com.mysql.cj.jdbc.Driver
    dataSource.setUrl(this.url); //that's the uri provided before
    dataSource.setUsername(USER);
    dataSource.setPassword(PASS);

    Connection conn = null;
    Statement stmnt = null;
    ResultSet dbs = null;
    try {       
        conn = dataSource.getConnection(); //And this is the line that triggers the exception.
        System.out.println(conn!=null);
        dbs = conn.getMetaData().getCatalogs();

        while(dbs.next() && !exists)
            if(dbs.getString(1).equals(dbName))
                exists = true;

    } catch (SQLException e) {
        e.printStackTrace();
    } finally {
        try {
            if(conn != null)
                conn.close();
            if(stmnt != null)
                stmnt.close();
            if(dbs != null)
                dbs.close();
            if(dataSource != null)
                dataSource.close();
        } catch (SQLException e) {
                logger.error(e.getMessage());
                return exists;
        }
    }

    return exists;
}

I tried the following things:

• From my local machine: telnet <public ip of the VM> 3306 -> Connection timed out
• From my local machine: mysql -h<public ip of the VM> -u user -p -> ERROR 2003 (HY000): Can't connect to MySQL server on 'remoteHostname.com' (110)
• From inside the VM: nc localhost 3306 -> I can get in.
• From inside the VM: mysql -h127.0.0.1 -u user -p -> I can get in.
• From inside the VM: mysql -h<public ip address> -u user -p -> I can get in.

If I run netstat -tulpn | grep 3306 i got:

tcp6       0      0 :::3306                 :::*                    LISTEN      -                   
tcp6       0      0 :::33060                :::*                    LISTEN      -  

This is my.cnf:

[mysqld]
default_authentication_plugin=mysql_native_password
datadir=/local/user/mysql
socket=/var/lib/mysql/mysql.sock
general_log_file=/var/log/mysql.log
general_log=1
log-error=/var/log/mysqld.log
pid-file=/var/run/mysqld/mysqld.pid

I can connect to the db through PhpMyAdmin, but I think this is due to the fact that's using a Socket connection.

If I check both /var/log/mysql.log and /var/log/mysqld.log (this is the one about the errors) there's no evidence of anything. Basically when I try to connect those don't display anything.

EDIT: If i use the same settings to connect to my local machine instance of mysql everything works fine. So could this be related to a networking problem? Or VM's instane of mysql doen't accept anymore tcp connection for some reason?

RE-EDIT: Checking in /var/log/messages I found out that the host machine went out of disk space. From that time on, the interface docker0 went on a blocked state and then disabled state loop.

Configuration:

• Remote VM with RHEL7
• 32 cores CPU
• 64GB RAM
• MySQL 8.0.13 - MySQL Community Server - GPL
• PhpMyAdmin 4.8.4

L. Don :

In the Re-edit section of the question i specified new details. I can't still run the application correctly with the old configuration but using --network=host parameter in docker run did the trick.

That's simply a workaround. I'll update this question with further informations.