In this section, we discuss how the MySQL privilege system works in relation to MySQL Cluster and the implications of this for keeping a MySQL Cluster secure.
Standard MySQL privileges apply to MySQL Cluster tables. This
includes all MySQL privilege types
DELETE privilege, and so on)
granted on the database, table, and column level. As with any
other MySQL Server, user and privilege information is stored in
mysql system database. The SQL statements
used to grant and revoke privileges on
NDB tables, databases containing
such tables, and columns within such tables are identical in all
respects with the
REVOKE statements used in
connection with database objects involving any (other) MySQL
It is important to keep in mind that the MySQL grant tables use
MyISAM storage engine. Because of this,
those tables are not duplicated or shared among MySQL servers
acting as SQL nodes in a MySQL Cluster. By way of example,
suppose that two SQL nodes A
and B are connected to the same
MySQL Cluster, which has an
mytable in a database named
mydb, and that you execute an SQL statement
on server A that creates a new
jon@localhost and grants this user the
SELECT privilege on that table:
GRANT SELECT ON mydb.mytable->
TO jon@localhost IDENTIFIED BY 'mypass';
In other words, changes in users and their privileges do not automatically propagate between SQL nodes. Synchronization of privileges between SQL nodes must be done either manually or by scripting an application that periodically synchronizes the privilege tables on all SQL nodes in the cluster.
Conversely, because there is no way in MySQL to deny privileges
(privileges can either be revoked or not granted in the first
place, but not denied as such), there is no special protection
NDB tables on one SQL node from
users that have privileges on another SQL node. The most
far-reaching example of this is the MySQL
root account, which can perform any action on
any database object. In combination with empty
config.ini file, this account can be
especially dangerous. To understand why, consider the following
config.ini file contains at least
[api] section. This means that the
Cluster management server performs no checking of the host
from which a MySQL Server (or other API node) accesses the
There is no firewall, or the firewall fails to protect against access to the Cluster from hosts external to the network.
The host name or IP address of the Cluster's management server is known or can be determined from outside the network.
In sum, you cannot have a safe MySQL Cluster if it is directly accessible from outside your local network.
Never leave the MySQL root account password empty. This is just as true when running MySQL as a MySQL Cluster SQL node as it is when running it as a standalone (non-Cluster) MySQL Server, and should be done as part of the MySQL installation process before configuring the MySQL Server as an SQL node in a MySQL Cluster.
You should never convert the system tables in the
mysql database to use the
NDB storage engine. There are a
number of reasons why you should not do this, but the most
important reason is this: Many of the SQL statements
mysql tables storing information
about user privileges, stored routines, scheduled events, and
other database objects cease to function if these tables are
changed to use any storage engine other than
MyISAM. This is a consequence of
various MySQL Server internals which are not expected to change
in the foreseeable future.
If you need to synchronize
tables between SQL nodes, you can use standard MySQL replication
to do so, or employ a script to copy table entries between the
Summary. The two most important points to remember regarding the MySQL privilege system with regard to MySQL Cluster are:
Users and privileges established on one SQL node do not automatically exist or take effect on other SQL nodes in the cluster.
Conversely, removing a user or privilege on one SQL node in the cluster does not remove the user or privilege from any other SQL nodes.
Once a MySQL user is granted privileges on an
NDB table from one SQL node
in a MySQL Cluster, that user can “see” any
data in that table regardless of the SQL node from which
the data originated.