MySQL-Performance-Tuner-Mcp

MySQL-Performance-Tuner-Mcp

MySQL MCP Performance Tuning Server - AI-powered MySQL performance tuning capabilities

Category
访问服务器

README

MySQL Performance Tuning MCP

A Model Context Protocol (MCP) server for MySQL performance tuning and analysis.

PyPI - Version PyPI - Downloads Python 3.10+ Pepy Total Downloads Docker Pulls

<a href="https://glama.ai/mcp/servers/@isdaniel/MySQL-Performance-Tuner-Mcp"> <img width="380" height="200" src="https://glama.ai/mcp/servers/@isdaniel/MySQL-Performance-Tuner-Mcp/badge" /> </a>

Overview

mysqltuner_mcp provides AI-powered MySQL database performance analysis through the Model Context Protocol. It offers tools for query optimization, index recommendations, health monitoring

Features

Performance Analysis

  • Slow Query Detection: Identify slow queries from performance_schema
  • Query Analysis: Get detailed EXPLAIN plans with recommendations
  • Table Statistics: Analyze table sizes, row counts, and fragmentation
  • Statement Analysis: Analyze SQL statements for temp tables, sorting, and full scans

Index Optimization

  • Index Recommendations: AI-powered suggestions based on query patterns
  • Unused Index Finder: Identify indexes that are never read
  • Duplicate Detection: Find redundant and overlapping indexes
  • Index Statistics: Cardinality, selectivity, and usage metrics

Health Monitoring

  • Health Check: Comprehensive database health assessment with scoring
  • Active Queries: Real-time query monitoring
  • Wait Event Analysis: Identify I/O and lock bottlenecks
  • Configuration Review: Settings analysis with recommendations

Storage Engine Analysis

  • Engine Statistics: Analyze storage engine usage and distribution
  • Fragmentation Detection: Find fragmented tables with OPTIMIZE recommendations
  • Auto-Increment Analysis: Detect columns approaching overflow limits

InnoDB Analysis

  • InnoDB Status: Parse and analyze SHOW ENGINE INNODB STATUS
  • Buffer Pool Analysis: Detailed buffer pool usage by schema and table
  • Transaction Analysis: Monitor transactions, lock waits, and deadlocks

Memory Analysis

  • Memory Calculations: Calculate per-thread and global buffer usage
  • Memory by Host/User: Breakdown memory usage by connection source
  • Table Cache Analysis: Analyze table open cache efficiency

Replication Monitoring

  • Master/Slave Status: Monitor replication health and lag
  • Galera Cluster: Full Galera cluster status for MariaDB/Percona
  • Group Replication: MySQL Group Replication monitoring

Security Analysis

  • Security Audit: Check for anonymous users, weak passwords, dangerous privileges
  • User Privileges: Analyze user privileges at all levels
  • Audit Log: Check audit logging configuration

Resources & Prompts

  • Built-in best practices documentation
  • Pre-configured prompts for common tuning tasks
  • Index optimization guidelines
  • Configuration optimization guide

Installation

From Source

git clone https://github.com/yourusername/mysqltuner_mcp.git
cd mysqltuner_mcp
pip install -e .

Using pip (when published)

pip install mysqltuner_mcp

Configuration

Environment Variables

Variable Description Default
MYSQL_URI MySQL connection URI (required) -
MYSQL_POOL_SIZE Connection pool size 5
MYSQL_SSL Enable SSL/TLS connection false
MYSQL_SSL_CA Path to CA certificate file -
MYSQL_SSL_CERT Path to client certificate file -
MYSQL_SSL_KEY Path to client private key file -
MYSQL_SSL_VERIFY_CERT Verify server certificate true
MYSQL_SSL_VERIFY_IDENTITY Verify server hostname matches certificate false

Connection URI Format

Environment Variables

export MYSQL_URI="mysql://user:password@host:3306/database"
export MYSQL_SSL=true
export MYSQL_SSL_CA="/path/to/ca.pem"  # Optional: CA certificate for verification

Connection URI Query Parameters

export MYSQL_URI="mysql://user:password@host:3306/database?ssl=true&ssl_ca=/path/to/ca.pem"

Usage

Running the Server

The server supports three transport modes: stdio (default), SSE, and streamable-http.

# As a module
python -m mysqltuner_mcp

# Using the entry point
mysqltuner-mcp

# Explicitly specifying stdio mode
python -m mysqltuner_mcp --mode stdio

SSE Mode (Server-Sent Events)

HTTP transport using Server-Sent Events, suitable for web-based MCP clients:

# Start SSE server on default port 8080
python -m mysqltuner_mcp --mode sse

# Specify custom host and port
python -m mysqltuner_mcp --mode sse --host 127.0.0.1 --port 3000

# Enable debug mode
python -m mysqltuner_mcp --mode sse --debug

SSE Endpoints:

  • http://<host>:<port>/sse - SSE connection endpoint
  • http://<host>:<port>/messages/ - Message posting endpoint

Streamable HTTP Mode

Modern HTTP transport with session management:

# Start streamable HTTP server (stateful, with session tracking)
python -m mysqltuner_mcp --mode streamable-http

# Start in stateless mode (fresh transport per request)
python -m mysqltuner_mcp --mode streamable-http --stateless

# Specify custom host and port
python -m mysqltuner_mcp --mode streamable-http --host 127.0.0.1 --port 3000

Streamable HTTP Endpoint:

  • http://<host>:<port>/mcp - Single endpoint for all MCP communication

Command-Line Options

Option Description Default
--mode Server mode: stdio, sse, or streamable-http stdio
--host Host to bind to (HTTP modes only) 0.0.0.0
--port Port to listen on (HTTP modes only) 8080 or PORT env var
--stateless Run in stateless mode (streamable-http only) false
--debug Enable debug logging false

MCP Client Configuration

Add to your MCP client configuration (e.g., Claude Desktop):

{
  "mcpServers": {
    "mysqltuner_mcp": {
      "command": "python",
      "args": ["-m", "mysqltuner_mcp"],
      "env": {
        "MYSQL_URI": "mysql://root:your_password@localhost:3306/your_database"
      }
    }
  }
}

With SSL/TLS Enabled

{
  "mcpServers": {
    "mysqltuner_mcp": {
      "command": "python",
      "args": ["-m", "mysqltuner_mcp"],
      "env": {
        "MYSQL_URI": "mysql://root:your_password@localhost:3306/your_database",
        "MYSQL_SSL": "true",
        "MYSQL_SSL_CA": "/path/to/ca.pem"
      }
    }
  }
}

Available Tools

Performance Tools

Tool Description
get_slow_queries Retrieve slow queries from performance_schema with detailed statistics
analyze_query Get EXPLAIN plan and analysis for a query with optimization recommendations
get_table_stats Get table statistics including size, row counts, fragmentation, and indexes

Index Tools

Tool Description
get_index_recommendations AI-powered index suggestions based on query patterns from performance_schema
find_unused_indexes Find unused, duplicate, and redundant indexes with DROP statements
get_index_stats Detailed index statistics including cardinality, selectivity, and usage metrics

Health Tools

Tool Description
check_database_health Comprehensive health check with scoring (connections, buffer pool, queries, etc.)
get_active_queries Monitor currently running queries and identify long-running/blocked queries
review_settings Analyze MySQL configuration settings with best practice recommendations
analyze_wait_events Identify wait event bottlenecks (I/O, locks, buffer, log waits)

Storage Engine Tools

Tool Description
analyze_storage_engines Analyze storage engine usage, statistics, and recommendations
get_fragmented_tables Find tables with significant fragmentation and wasted space
analyze_auto_increment Check auto-increment columns for potential overflow issues

InnoDB Tools

Tool Description
get_innodb_status Parse and analyze SHOW ENGINE INNODB STATUS output
analyze_buffer_pool Detailed InnoDB buffer pool analysis by schema and table
analyze_innodb_transactions Analyze InnoDB transactions, lock waits, and deadlocks

Memory Tools

Tool Description
calculate_memory_usage Calculate MySQL memory usage (per-thread and global buffers)
get_memory_by_host Get memory usage breakdown by host, user, or event
get_table_memory_usage Analyze table cache and InnoDB buffer pool by table

Replication Tools

Tool Description
get_replication_status Get master/slave replication status and health
get_galera_status Get Galera cluster status (MariaDB/Percona XtraDB Cluster)
get_group_replication_status Get MySQL Group Replication status

Security Tools

Tool Description
analyze_security Comprehensive security analysis (users, passwords, SSL, privileges)
analyze_user_privileges Analyze privileges for specific users or all users
check_audit_log Check audit log configuration and status

Statement Analysis Tools

Tool Description
analyze_statements Comprehensive SQL statement analysis from performance_schema
get_statements_with_temp_tables Find statements creating temporary tables (memory and disk)
get_statements_with_sorting Find statements with sorting operations and file sorts
get_statements_with_full_scans Find statements performing full table scans
get_statements_with_errors Find statements producing errors or warnings

Available Prompts

Prompt Description
optimize_slow_query Analyze and optimize a slow query
health_check Perform comprehensive health assessment
index_review Review indexes for a database
performance_audit Full performance audit

Requirements

  • Python 3.10+
  • MySQL 5.7+ or MySQL 8.0+
  • performance_schema enabled (for full functionality)

MySQL Permissions

The MySQL user needs the following privileges:

GRANT SELECT ON performance_schema.* TO 'your_user'@'%';
GRANT SELECT ON information_schema.* TO 'your_user'@'%';
GRANT PROCESS ON *.* TO 'your_user'@'%';
-- For EXPLAIN on user databases:
GRANT SELECT ON your_database.* TO 'your_user'@'%';

Development

Setup Development Environment

git clone https://github.com/yourusername/mysqltuner_mcp.git
cd mysqltuner_mcp
python -m venv .venv
source .venv/bin/activate  # or .venv\Scripts\activate on Windows
pip install -e .

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选