How to Configure Database Connection Pooling with Druid in the Mall Project

The Mall project configures database connection pooling with Alibaba Druid using Spring Boot YAML properties under the spring.datasource.druid namespace, enabling high-performance database connectivity with built-in monitoring across all microservices.

The macrozheng/mall repository demonstrates enterprise-grade database connection management by integrating Alibaba Druid as its JDBC connection pool. This approach leverages Spring Boot auto-configuration to eliminate boilerplate code while providing production-ready monitoring capabilities through the Druid console.

Adding the Druid Dependency

Druid is introduced to the project through the druid-spring-boot-starter dependency defined in the root pom.xml. The repository pins version 1.2.14 in the dependency management section, ensuring all service modules—including mall-admin, mall-portal, and mall-search—inherit the same consistent version.

<!-- In pom.xml (root) -->
<dependencyManagement>
  <dependencies>
    <dependency>
      <groupId>com.alibaba</groupId>
      <artifactId>druid-spring-boot-starter</artifactId>
      <version>1.2.14</version>
    </dependency>
  </dependencies>
</dependencyManagement>

With the starter on the classpath, Spring Boot automatically creates a DruidDataSource bean without requiring explicit Java configuration, unless custom filters or listeners are needed.

Configuring Connection Pool Settings via YAML

Each service module ships with environment-specific YAML configurations (application-dev.yml and application-prod.yml) that define the connection pool parameters under the spring.datasource.druid key.

In mall-admin/src/main/resources/application-dev.yml, the development profile specifies:

spring:
  datasource:
    url: jdbc:mysql://localhost:3306/mall?useUnicode=true&characterEncoding=utf-8&serverTimezone=Asia/Shanghai&useSSL=false
    username: root
    password: root
    druid:
      initial-size: 5          # initial connections

      min-idle: 10             # minimum idle connections

      max-active: 20           # maximum active connections

      web-stat-filter:
        exclusions: "*.js,*.gif,*.jpg,*.png,*.css,*.ico,/druid/*"
      stat-view-servlet:
        login-username: druid
        login-password: druid

The initial-size parameter establishes five connections on startup, while min-idle maintains ten idle connections and max-active caps the total at twenty concurrent connections. These values should be tuned according to your specific workload and database server capacity.

The production configuration in mall-admin/src/main/resources/application-prod.yml mirrors this structure with hardened credentials and host values appropriate for production environments.

Enabling the Druid Monitoring Dashboard

Druid provides a built-in web interface for real-time connection pool statistics, SQL profiling, and wall filters. This is enabled through the web-stat-filter and stat-view-servlet properties shown above.

The monitoring console is accessible at:

http://<host>:<port>/druid/index.html

Use the credentials defined in stat-view-servlet (default: druid / druid) to authenticate.

To ensure the monitoring endpoint remains accessible, the global security configuration explicitly excludes the /druid/** path from authentication filters. In mall-admin/src/main/resources/application.yml, the security whitelist permits unauthenticated access to the Druid console:

spring:
  security:
    ignored:
      - /druid/**

Maintaining Cross-Module Consistency

The Mall project enforces uniform database connection behavior across all microservices by replicating the same Druid configuration pattern. The mall-portal module defines identical pool settings in mall-portal/src/main/resources/application-dev.yml, while mall-search implements the same structure in mall-search/src/main/resources/application-dev.yml.

This consistency ensures that whether a request hits the admin interface, user portal, or search service, the database connection pooling characteristics remain predictable and manageable.

Customizing Druid Programmatically

While YAML configuration handles most scenarios, you can programmatically customize the DruidDataSource when requiring advanced filters like SQL wall protection or custom stat logging.

Create a configuration class to override the auto-configured bean:

import com.alibaba.druid.pool.DruidDataSource;
import com.alibaba.druid.filter.stat.StatFilter;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import java.util.Collections;

@Configuration
public class DruidConfig {

    @Bean
    public DruidDataSource druidDataSource() {
        DruidDataSource ds = new DruidDataSource();
        // Basic properties still load from application.yml
        StatFilter statFilter = new StatFilter();
        statFilter.setMergeSql(true);
        ds.setProxyFilters(Collections.singletonList(statFilter));
        return ds;
    }
}

This approach allows you to inject custom filters while retaining the YAML-based connection pool sizing.

Summary

• Dependency Setup: Add druid-spring-boot-starter version 1.2.14 to the root pom.xml to enable Druid across all modules.
• YAML Configuration: Define pool sizing (initial-size, min-idle, max-active) under spring.datasource.druid in each module's application-dev.yml or application-prod.yml.
• Monitoring Access: Enable the web console via stat-view-servlet properties and whitelist /druid/** in security configuration to access real-time metrics.
• Consistency: Replicate the same Druid configuration structure across mall-admin, mall-portal, and mall-search for uniform behavior.
• Programmatic Extension: Override the DruidDataSource bean with custom filters when YAML configuration proves insufficient.

Frequently Asked Questions

How do I change the maximum number of database connections in the Mall project?

Modify the max-active property under spring.datasource.druid in your active profile's YAML file (e.g., mall-admin/src/main/resources/application-dev.yml). The default development configuration sets this value to 20, but you should increase it for high-traffic environments or decrease it to conserve database server resources.

Where is the Druid monitoring console URL in the Mall application?

Access the monitoring dashboard at http://<host>:<port>/druid/index.html after starting any service module. The default login credentials are druid for both username and password, as configured in the stat-view-servlet section of the YAML files. Ensure the /druid/** path is excluded from Spring Security filters to avoid authentication redirects.

Why does the Mall project use Druid instead of HikariCP?

According to the macrozheng/mall source code, Druid is chosen for its built-in monitoring capabilities (the /druid web console), configurable filters (web-stat, wall, and log filters via YAML), and high-performance connection acquisition. While HikariCP offers faster raw performance, Druid provides superior observability and SQL profiling features essential for production debugging without additional tooling.

Can I use a custom password for the Druid monitoring login?

Yes. Change the login-username and login-password values under stat-view-servlet in your application.yml or profile-specific configuration files. For production deployments in application-prod.yml, you should always replace the default druid/druid credentials with secure, environment-specific values before deploying to production servers.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}