Technology sharing | MySQL shell customized deployment MySQL instance

author ： Yang Taotao

Senior database expert , Specialized research MySQL More than ten years . Good at MySQL、PostgreSQL、MongoDB And other open source database related backup recovery 、SQL tuning 、 Monitoring operation and maintenance 、 High availability architecture design, etc . At present, I work in akerson , For major operators and bank financial enterprises MySQL Related technical support 、MySQL Related courses, training, etc .

In this paper, the source ： Original contribution

* Produced by aikesheng open source community , Original content is not allowed to be used without authorization , For reprint, please contact the editor and indicate the source .

---

Previously written and used MySQL Shell Of DBA Components to build 、 Operation and maintenance MySQL InnoDB Cluster . Today, let's continue on DBA Extended introduction of components , Discuss how to use it to customize the deployment MySQL example .

One 、 How to deploy multiple versions MySQL example ：

By default DBA When a component deploys multiple instances , Using the built-in MySQL edition , That is, in the environment variable $PATH Search in the included path mysqld To confirm the specific instance version . For example, on my machine , Default MySQL Version is 8.0 , So the instance version after direct deployment is also 8.0.

[email protected]:~# mysqlsh --py 
MySQL Shell 8.0.29
   
...
MySQL  Py > dba.deploy_sandbox_instance(3350);
A new MySQL sandbox instance will be created on this host in 
/root/mysql-sandboxes/3350
   
...
   
Instance localhost:3350 successfully deployed and started.
Use shell.connect('[email protected]:3350') to connect to the instance.

For example, the above port is 3350 After the deployment is completed , Create a new connection , After success, the client will receive the specific version ：Server version: 8.0.29 MySQL Community Server - GPL.

MySQL  Py > \c [email protected]:3350
...
Your MySQL connection id is 12
Server version: 8.0.29 MySQL Community Server - GPL

If you want to deploy multiple different versions at the same time MySQL example , Just put the corresponding version of mysqld Path into $ PATH that will do . It should be noted that , New path and old $ PATH Splicing sequence ： The new path comes first , The old $ PATH After ！ After deployment , Remember to put it again $PATH Content restore .

For example, on my machine MySQL 5.7 The directory of the version installation package is ：/root/opt/mysql/5.7.34 , Then add subdirectories of this directory bin To environment variable $PATH that will do ：

[email protected]:~# export PATH=/root/opt/mysql/5.7.34/bin:$PATH

Re enter MySQL Shell Environmental Science , And above MySQL 8.0 The same deployment method ： You can see , The newly deployed instance version is 5.7.34-log MySQL Community Server (GPL).

MySQL  Py > dba.deploy_sandbox_instance(3351)
A new MySQL sandbox instance will be created on this host in 
/root/mysql-sandboxes/3351
...
   
Instance localhost:3351 successfully deployed and started.
Use shell.connect('[email protected]:3351') to connect to the instance.
   
MySQL  Py > \c [email protected]:3351
...
Your MySQL connection id is 7
Server version: 5.7.34-log MySQL Community Server (GPL)

Two 、 How to change the basic directory of a deployment instance ：

The default deployment instance file is ~/mysql-sandboxes Next , Divide by instance port , One subdirectory per port . For example, the two previously deployed MySQL example , Corresponding to the directory /root/mysql-sandboxes/3350、/root/mysql-sandboxes/3351 .

There are two ways to change the basic directory of the deployment instance ：

1. call dba.deploy_sandbox_instance when , Explicitly specify the deployment directory ：

- sandboxDir: path where the new instance will be deployed.

For example, deploy a new instance 3352, Specify the basic directory as ：/tmp/mysql-sandbox.

MySQL  Py > dba.deploy_sandbox_instance(3352,{"sandboxDir":"/tmp/mysql-sandbox"})
A new MySQL sandbox instance will be created on this host in 
 /tmp/mysql-sandbox/3352
      
...
Instance localhost:3352 successfully deployed and started.
Use shell.connect('[email protected]:3352') to connect to the instance.

The biggest disadvantage of this method is that it is not universal for the deployment of subsequent new instances , You need to specify... For each new instance sandboxDir Option . If you don't explicitly specify , Then continue to use the default directory ：~/mysql-sandboxes. For example, the following deployment example 3353, Still use the default directory .

MySQL  Py > dba.deploy_sandbox_instance(3353)
A new MySQL sandbox instance will be created on this host in 
/root/mysql-sandboxes/3353

The next way is directly in MySQL Shell Of Shell Specify the basic deployment directory in the component , It works for the overall situation .

2. Set explicitly shell Component's options Dictionary properties , modify KEY be known as sandboxDir The value of is the specified directory ：

- sandboxDir: default path where the new sandbox instances for InnoDB cluster will be deployed

Set up sandboxDir by /tmp/mysql-sandbox ： --persist For permanent effect , similar MySQL Inside set persist Grammatical effect .

MySQL  Py > \option --persist sandboxDir /tmp/mysql-sandbox

Re enter MySQL Shell Environmental Science , Deploy two new instances , The corresponding ports are 3353 and 3354： Both instances are deployed in the directory /tmp/mysql-sandbox Next .

MySQL  Py > for i in range(3353,3355):
         ->     dba.deploy_sandbox_instance(i)
         -> 
A new MySQL sandbox instance will be created on this host in 
/tmp/mysql-sandbox/3353
...
A new MySQL sandbox instance will be created on this host in 
/tmp/mysql-sandbox/3354
...

3、 ... and 、 How to change the instance parameters of a new deployment ：

No specific parameters are set for the above deployed instances , Only the default values are used . No matter what scenario the deployed instance is used in , Some parameters still need to change with the scene . There are two ways to change parameters ：

1. Configure parameters while deploying instances ： Suitable for changing a few parameters .

For example, deploy a new instance 3355, Specify the following parameters respectively ：

server-id=1000

tmp_table_size=64M

read_buffer_size=1M

Add these parameters to mysqldOptions The array can be .

MySQL  Py > dba.deploy_sandbox_instance(3355,{"mysqldOptions":["server_id=1000","tmp_table_size=64M","read_buffer_size=1M"]})
A new MySQL sandbox instance will be created on this host in 
/tmp/mysql-sandbox/3355
  
...
      
Instance localhost:3355 successfully deployed and started.
Use shell.connect('[email protected]:3355') to connect to the instance.
      

confirm my.cnf Successfully changed ：

      [email protected]:/tmp/mysql-sandbox/3355# grep "server_id\|tmp_table_size\|read_buffer_size" my.cnf
      server_id = 1000
      tmp_table_size = 64M
      read_buffer_size = 1M

2. Configure the parameters after deploying the instance ： Suitable for changing a large number of parameters .

Directly change the corresponding my.cnf , After that, restart .

summary ：

Use MySQL Shell Of DBA Components can be deployed easily MySQL example , But it is not recommended to produce , Because of entering MySQL Shell After environment , These instances can be maintained at will , Increased risk .