pgsql

PgSQL Database Plugin

This is a PgSQL database connection plugin for the Lynx framework, providing complete database connection management functionality.

Features

✅ Implemented Features

1. Configuration Validation: Automatic validation of configuration parameter validity
2. Error Handling: Graceful error handling to avoid panics
3. Retry Mechanism: Automatic retry on connection failure
4. Connection Pool Monitoring: Real-time monitoring of connection pool status
5. Health Check: Comprehensive database health check
6. Graceful Shutdown: Safe connection closing mechanism
7. Configuration Updates: Support for runtime configuration updates
8. Detailed Logging: Provides detailed debugging and monitoring logs
9. Statistics: Provides connection pool statistics
10. Status Query: Provides connection status query interface

Configuration Guide

Basic Configuration

lynx:
  pgsql:
    driver: "postgres"
    source: "postgres://username:password@host:port/database?sslmode=disable"
    min_conn: 10
    max_conn: 50
    max_idle_time: "30s"
    max_life_time: "300s"

Configuration Parameters

Connection String Format

postgres://username:password@host:port/database?param1=value1&param2=value2

Common parameters:

• sslmode: SSL mode (disable, require, verify-ca, verify-full)
• connect_timeout: Connection timeout
• statement_timeout: Statement timeout
• application_name: Application name

Usage

1. Getting Database Driver

import (
    "github.com/go-lynx/lynx/plugins/db/pgsql"
    "entgo.io/ent/dialect/sql"
)

// Get database driver
driver := pgsql.GetDriver()
if driver == nil {
    // Handle error
    return
}

// Create client using ent
client := ent.NewClient(ent.Driver(driver))

2. Health Check

// Perform health check
err := pgsql.CheckHealth()
if err != nil {
    log.Errorf("Database health check failed: %v", err)
}

3. Getting Connection Pool Statistics

// Get connection pool statistics
stats := pgsql.GetStats()
if stats != nil {
    log.Infof("Connection pool stats: open=%d, in_use=%d, idle=%d", 
        stats.OpenConnections, stats.InUse, stats.Idle)
}

4. Checking Connection Status

// Check if connected
if pgsql.IsConnected() {
    log.Info("Database is connected")
} else {
    log.Warn("Database is not connected")
}

5. Getting Configuration Information

// Get current configuration
config := pgsql.GetConfig()
if config != nil {
    log.Infof("Current config: driver=%s, max_conn=%d", 
        config.Driver, config.MaxConn)
}

6. Prometheus Monitoring

// Get Prometheus metrics handler
handler := pgsql.GetPrometheusHandler()

// Register metrics endpoint in HTTP server
http.Handle("/metrics", handler)

Prometheus Monitoring Configuration

Enabling Monitoring

Enable Prometheus monitoring in the configuration file:

lynx:
  pgsql:
    driver: "postgres"
    source: "postgres://user:pass@localhost:5432/dbname"
    min_conn: 10
    max_conn: 50
    prometheus:
      enabled: true
      metrics_path: "/metrics"
      metrics_port: 9090
      namespace: "lynx"
      subsystem: "pgsql"
      labels:
        environment: "production"
        service: "myapp"

Monitoring Metrics

The plugin provides the following Prometheus metrics:

Connection Pool Metrics

• lynx_pgsql_max_open_connections: Maximum number of connections
• lynx_pgsql_open_connections: Current number of connections
• lynx_pgsql_in_use_connections: Number of connections in use
• lynx_pgsql_idle_connections: Number of idle connections
• lynx_pgsql_max_idle_connections: Maximum number of idle connections

Wait Metrics

• lynx_pgsql_wait_count_total: Total number of connection waits
• lynx_pgsql_wait_duration_seconds_total: Total time waiting for connections

Connection Close Metrics

• lynx_pgsql_max_idle_closed_total: Number of connections closed due to idle timeout
• lynx_pgsql_max_lifetime_closed_total: Number of connections closed due to lifetime expiration

Health Check Metrics

• lynx_pgsql_health_check_total: Total number of health checks
• lynx_pgsql_health_check_success_total: Number of successful health checks
• lynx_pgsql_health_check_failure_total: Number of failed health checks

Configuration Metrics

• lynx_pgsql_config_min_connections: Configured minimum number of connections
• lynx_pgsql_config_max_connections: Configured maximum number of connections

Accessing Monitoring Metrics

After starting the application, you can access monitoring metrics in the following ways:

# Access metrics endpoint
curl http://localhost:9090/metrics

# View specific metrics
curl http://localhost:9090/metrics | grep lynx_pgsql

Prometheus Configuration

Add scrape targets in the Prometheus configuration file:

scrape_configs:
  - job_name: 'lynx-pgsql'
    static_configs:
      - targets: ['localhost:9090']
    metrics_path: '/metrics'
    scrape_interval: 15s

Grafana Dashboard

You can create Grafana dashboards to visualize monitoring metrics:

{
  "dashboard": {
    "title": "Lynx PgSQL Monitoring",
    "panels": [
      {
        "title": "Connection Pool Status",
        "type": "stat",
        "targets": [
          {
            "expr": "lynx_pgsql_open_connections",
            "legendFormat": "Current Connections"
          }
        ]
      },
      {
        "title": "Connection Pool Utilization",
        "type": "gauge",
        "targets": [
          {
            "expr": "lynx_pgsql_in_use_connections / lynx_pgsql_max_open_connections * 100",
            "legendFormat": "Utilization %"
          }
        ]
      }
    ]
  }
}

Monitoring and Debugging

Connection Pool Statistics

The plugin provides the following statistics:

• MaxOpenConnections: Maximum open connections
• OpenConnections: Current open connections
• InUse: Connections in use
• Idle: Idle connections
• MaxIdleConnections: Maximum idle connections
• WaitCount: Number of connection waits
• WaitDuration: Total time waiting for connections
• MaxIdleClosed: Number of connections closed due to idle timeout
• MaxLifetimeClosed: Number of connections closed due to lifetime expiration

Log Information

The plugin outputs detailed log information:

• Configuration loading and validation
• Connection establishment process
• Retry attempts
• Health check results
• Connection pool status
• Error and warning messages

Error Handling

The plugin implements comprehensive error handling mechanisms:

1. Configuration Validation Errors: Validate configuration validity during initialization
2. Connection Errors: Automatic retry on connection failure
3. Health Check Errors: Provide detailed health check error information
4. Shutdown Errors: Gracefully handle connection closing errors

Best Practices

1. Connection Pool Configuration

# Development environment
min_conn: 5
max_conn: 20

# Production environment
min_conn: 20
max_conn: 200

2. Timeout Configuration

# Reasonable timeout configuration
max_idle_time: "300s"    # 5 minutes
max_life_time: "3600s"   # 1 hour

3. SSL Configuration

# Development environment
source: "postgres://user:pass@localhost:5432/db?sslmode=disable"

# Production environment
source: "postgres://user:pass@db.example.com:5432/db?sslmode=require"

4. Monitoring Integration

// Periodically check connection pool status
go func() {
    ticker := time.NewTicker(30 * time.Second)
    defer ticker.Stop()
    
    for {
        select {
        case <-ticker.C:
            stats := pgsql.GetStats()
            if stats != nil {
                // Send monitoring metrics
                metrics.RecordConnectionPoolStats(stats)
            }
        }
    }
}()

Troubleshooting

Common Issues

1. Connection Failure

   • Check connection string format
   • Verify database service is running
   • Check network connection

2. Connection Pool Exhaustion

   • Increase max_conn configuration
   • Check for connection leaks
   • Optimize query performance

3. Health Check Failure

   • Check database service status
   • Verify network connection
   • Check detailed error logs

Debugging Tips

1. Enable detailed logging
2. Monitor connection pool statistics
3. Perform regular health checks
4. Use connection pool monitoring tools

Version History

• v2.0.0: Refactored version, added complete error handling, retry mechanism, monitoring functions, etc.
• v1.x.x: Basic version, provides fundamental database connection functionality