In addition to Weibo, there is also WeChat
Please pay attention
WeChat public account
Shulou
2025-04-02 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > Database >
Share
Shulou(Shulou.com)06/01 Report--
1. The role of MySQL cluster:
- Resolving single points of failure of access nodes
- Single point of failure of data storage nodes
- Solve data storage node data backup problems
2. Cluster:
Use a set of servers to provide the same service
About MySQL-Cluster:
MySQL official cluster version
Integrated standard MySQL program, can be installed independently
Using NDB (Network DataBase) engine
Assume that each node has its own memory, hard disk,
Reduce single points of failure of the entire database with inexpensive hardware
The role of the server in the cluster
- Data node: ndbd(single thread) ndb_mtd(multithread) stores data (records in table)
- SQL node: mysqld is the interface for clients to access data, responsible for executing SQL commands
It can be understood as a standard MySQL server supporting NDB, only storing table structure, independent user authorization
Provide access services such as user authentication and SQL query for clients
- Management node: ndb_mgmd manages all servers in the cluster
Client: client accessing data
5. Case topology:
6. MySQL cluster setup (configuration)
1) Common configuration:
# rpm -qa | grep -i mysql
# service mysql stop ; chkconfig mysql off
# rpm -e --nodeps MySQL-devel MySQL-embeddedMySQL-test MySQL-server MySQL-client MySQL-shared MySQL-shared-compat
//Keep RHEL's own mysql-libs package
# rm -rf /etc/my.cnf
# rm -rf /var/lib/mysql/*
# tar -xvfMySQL-Cluster-gpl-7.3.3-1.el6.x86_64.rpm-bundle.tar
# rpm -Uvh MySQL-Cluster-*.rpm
2) Configuration management node:
Run the admin process ndb_mgmd
The requested URL/etc/config.ini was not found on this server.
[ndbd default]: Common configuration of the data node
[ndb_mgmd]: Specify an administrative node
[ndbd]: Specify the data node
[mysqld]: Specify sql node
# mkdir -p /var/log/mysql-cluster //Create working directory
# vim/etc/config. ini//Write the main configuration file
[ndbd default]
NoOfReplicas=2 //Keep 2 copies of data
DataMemory=80M //Data cache size
IndexMemory=18M //Index cache size
[ndb_mgmd]
nodeid=7 //ID of the management node
hostname= 192.168.4.100//IP address of management node
datadir=/var/log/mysql-cluster //working directory
[ndbd] //Set data node ndbA
nodeid=8 //ID of data node ndbA
hostname= 192.168.4.30//IP address of this node
datadir=/var/log/mysql-cluster/data //working directory
[ndbd] //Set data node ndbB
nodeid=9
hostname=192.168.4.40
datadir=/var/log/mysql-cluster/data
[mysqld] //Set sql node sqlA
nodeid=10 //sql ID ID of node sqlA
hostname= 192.168.4.10//IP address of this node
[mysqld] //Set sql node sqlB
nodeid=11
hostname=192.168.4.20
:wq
3) Configure data nodes:
Run data process ndbd
Configuration file: /etc/my.cnf
#mkdir -p /var/log/mysql-cluster/data //Create working directory
#vim /etc/my.cnf
[mysqld]
datadir=/var/log/mysql-cluster/data //working directory
ndb-connectstring= 192.168.4.100//IP address of management node
ndbcluster //Specifies the use of the ndbcluster cluster storage engine
[mysql_cluster]
ndb-connectstring= 192.168.4.100//How to connect management nodes
:wq
4) Configure SQL node
Run mysql service mysql (this service is provided by mysql-cluster package)
#vim /etc/my.cnf
[mysqld]
ndbcluster //Specifies the use of the ndbcluster cluster storage engine
default-storage-engine=ndbcluster //specifies that the default storage engine for tables is ndbcluster
ndb-connectstring=192.168.4.100
[mysql_cluster]
ndb-connectstring= 192.168.4.100//How to connect management nodes
:wq
7. Start processes on different role servers
1) Start the management process on the management node
#ndb_mgmd -f /etc/config.ini
//Start the process, read the.ini configuration with the-f option; default background mode, add--nodaemon when debugging; after modifying the configuration, add--initial to reinitialize
#netstat -untlap |grep :1186 //View progress
#pkill -9 ndb_mgmd //kill process
2) Start the data process on the data node in turn
#ndbd
//When executed for the first time, add--initial initialization; multiple data nodes operate the same
3) Start database services on sql node in turn
#service mysql start //multiple sql nodes operate the same
8. View cluster status
1) Log in to the management interface on the management node to view the status.
#ndb_mgm
ndb_mgm> show
Cluster Configuration
---------------------
[ndbd(NDB)] 2node(s)
id=8 @192.168.4.30 (mysql-5.6.14 ndb-7.3.3, Nodegroup: 0, *)
id=9 @192.168.4.40 (mysql-5.6.14 ndb-7.3.3, Nodegroup: 0)
[ndb_mgmd(MGM)] 1 node(s)
id=7 @192.168.4.100 (mysql-5.6.14 ndb-7.3.3)
[mysqld(API)] 2 node(s)
id=10 @192.168.4.10 (mysql-5.6.14 ndb-7.3.3)
id=11 @192.168.4.20 (mysql-5.6.14 ndb-7.3.3)
ndb_mgm>
2) MySQL initialization of Sql node
Reset root password; add user authorization for client access
#service mysqlstart
#mysql -uroot -p initial password//initial password see/root/.mysql_secret file
3) On SQL node, check default storage engine
Mysql>show engines;
9. Test access nodes for single point failures
Authorization on sql nodes allows you to connect to yourself from remote clients
Grant all on webdb.* to webuser@"%"identified by "webuser88"; //both sql nodes do authorization
Client access: #mysql-h292.168.10/20 -uwebuser-pwebuser88
mysql>create database webdb;
mysql>create table webdb.a(id int);
mysql>insert into webdb.a values(1000);
mysql> select * from webdb.a ;
Access sqlA node manipulates data and synchronizes results on sqlB.
10. Single point failure of test data node
Data nodes can backup each other, automatically elect masters, and automatically synchronize data. During testing, you can turn off the ndbd service of a data node, insert and query data to verify the synchronization of data.
After the process of SQL node and data node runs normally, if there is no need to restart, the management node can be shut down.
Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.
Views: 0
*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.
Continue with the installation of the previous hadoop.First, install zookooper1. Decompress zookoope
"Every 5-10 years, there's a rare product, a really special, very unusual product that's the most un
© 2024 shulou.com SLNews company. All rights reserved.