- 22.214.171.124 NDB Cluster Configuration: Basic Example
- 126.96.36.199 Recommended Starting Configuration for NDB Cluster
- 188.8.131.52 NDB Cluster Connection Strings
- 184.108.40.206 Defining Computers in an NDB Cluster
- 220.127.116.11 Defining an NDB Cluster Management Server
- 18.104.22.168 Defining NDB Cluster Data Nodes
- 22.214.171.124 Defining SQL and Other API Nodes in an NDB Cluster
- 126.96.36.199 Defining the System
- 188.8.131.52 MySQL Server Options and Variables for NDB Cluster
- 184.108.40.206 NDB Cluster TCP/IP Connections
- 220.127.116.11 NDB Cluster TCP/IP Connections Using Direct Connections
- 18.104.22.168 NDB Cluster Shared-Memory Connections
- 22.214.171.124 Data Node Memory Management
- 126.96.36.199 Configuring NDB Cluster Send Buffer Parameters
Configuring NDB Cluster requires working with two files:
my.cnf: Specifies options for all NDB Cluster executables. This file, with which you should be familiar with from previous work with MySQL, must be accessible by each executable running in the cluster.
config.ini: This file, sometimes known as the global configuration file, is read only by the NDB Cluster management server, which then distributes the information contained therein to all processes participating in the cluster.
config.inicontains a description of each node involved in the cluster. This includes configuration parameters for data nodes and configuration parameters for connections between all nodes in the cluster. For a quick reference to the sections that can appear in this file, and what sorts of configuration parameters may be placed in each section, see Sections of the
Caching of configuration data.
NDB uses stateful
configuration. Rather than reading the global
configuration file every time the management server is
restarted, the management server caches the configuration the
first time it is started, and thereafter, the global
configuration file is read only when one of the following
conditions is true:
The management server is started using the --initial option. When
--initialis used, the global configuration file is re-read, any existing cache files are deleted, and the management server creates a new configuration cache.
The management server is started using the --reload option. The
--reloadoption causes the management server to compare its cache with the global configuration file. If they differ, the management server creates a new configuration cache; any existing configuration cache is preserved, but not used. If the management server's cache and the global configuration file contain the same configuration data, then the existing cache is used, and no new cache is created.
The management server is started using --config-cache=FALSE. This disables
--config-cache(enabled by default), and can be used to force the management server to bypass configuration caching altogether. In this case, the management server ignores any configuration files that may be present, always reading its configuration data from the
No configuration cache is found. In this case, the management server reads the global configuration file and creates a cache containing the same configuration data as found in the file.
Configuration cache files.
The management server by default creates configuration cache
files in a directory named
the MySQL installation directory. (If you build NDB Cluster from
source on a Unix system, the default location is
/usr/local/mysql-cluster.) This can be
overridden at runtime by starting the management server with the
Configuration cache files are binary files named according to
node_id is the management
server's node ID in the cluster, and
seq_id is a cache idenitifer. Cache
files are numbered sequentially using
seq_id, in the order in which they
are created. The management server uses the latest cache file as
determined by the
It is possible to roll back to a previous configuration by
deleting later configuration cache files, or by renaming an
earlier cache file so that it has a higher
seq_id. However, since configuration
cache files are written in a binary format, you should not
attempt to edit their contents by hand.
For more information about the
--reload options for the NDB
Cluster management server, see
Section 22.4.4, “ndb_mgmd — The NDB Cluster Management Server Daemon”.
We are continuously making improvements in Cluster configuration and attempting to simplify this process. Although we strive to maintain backward compatibility, there may be times when introduce an incompatible change. In such cases we will try to let Cluster users know in advance if a change is not backward compatible. If you find such a change and we have not documented it, please report it in the MySQL bugs database using the instructions given in Section 1.7, “How to Report Bugs or Problems”.