linghengqian
commented
2 years ago Actually, I thought about a similar solution. Referring to the processing ideas of https://github.com/baomidou/dynamic-datasource-spring-boot-starter , I think it is possible to add the global configuration Map of hikaricp, dbcp and tomcat dbcp under spring.datasource.shardingsphere (with these JDBC pool is used as an example because the source code of shardingsphere imports these three components as the implementation of the metadata spi). The data source instance can automatically extends these automatic configurations, and the configuration in the data source instance will override the globally configured JDBC pool properties. But the shardingsphere proxy is being retrofitted with vert.x client, which makes me wonder if it will be too complicated.
spring:
datasource:
dynamic:
hikari:
catalog:
connection-timeout:
validation-timeout:
idle-timeout:
leak-detection-threshold:
max-lifetime:
max-pool-size:
min-idle:
initialization-fail-timeout:
connection-init-sql:
connection-test-query:
dataSource-class-name:
dataSource-jndi-name:
schema:
transaction-isolation-name:
is-auto-commit:
is-read-only:
is-isolate-internal-queries:
is-register-mbeans:
is-allow-pool-suspension:
data-source-properties:
health-check-properties:
datasource:
master:
username: root
password: 123456
driver-class-name: com.mysql.jdbc.Driver
url: jdbc:mysql://xx.xx.xx.xx:3306/dynamic?characterEncoding=utf8&useSSL=false
hikari:
max-pool-size:
idle-timeout:
huntkalio
Problem description: Multiple data sources use the same database instance. For example, if the host is 127.0.0.1 and the port is 3306, multiple data sources need to be created. If the same database instance has 100 databases, 100 data sources need to be created.
Solution: The data source connection does not specify a specific database, the data source is only a database connection, and the real database corresponding to the data source is maintained through memory map mapping.
Add map in memory for data source and database mapping relationship