MySQL故障排查与生产环境优化

复制代码

故障排查流程：
┌─────────────────────────────────────────────────────────┐
│ 1. 现象确认 → 2. 信息收集 → 3. 问题定位 → 4. 原因分析 │
│                              ↓                         │
│ 5. 解决方案 → 6. 验证效果 → 7. 总结复盘 → 8. 预防措施 │
└─────────────────────────────────────────────────────────┘

故障类型：
├── 连接类故障
├── 性能类故障
├── 数据类故障
├── 复制类故障
├── 资源类故障
└── 安全类故障

复制代码

# 1. 系统层面工具
top          # CPU、内存使用情况
htop         # 更友好的top
vmstat 1     # 虚拟内存统计
iostat -x 1  # 磁盘I/O统计
netstat -an  # 网络连接状态
ss -tlnp     # socket统计
lsof         # 打开文件列表

# 2. MySQL工具
mysqladmin   # 管理命令
mysqldump    # 备份工具
mysqlbinlog  # 二进制日志查看
pt-query-digest  # Percona工具包
mysqltuner   # 配置调优建议

# 3. 性能分析工具
perf         # 性能分析
strace       # 系统调用跟踪
tcpdump      # 网络抓包

复制代码

# MySQL日志位置
/var/log/mysql/error.log      # 错误日志
/var/log/mysql/slow.log       # 慢查询日志
/var/log/mysql/mysql-bin.*    # 二进制日志
/var/lib/mysql/*.err          # 错误日志（某些版本）

# 查看日志配置
mysql> SHOW VARIABLES LIKE '%log%';

# 查看错误日志
tail -100 /var/log/mysql/error.log

# 查看慢查询日志
tail -100 /var/log/mysql/slow.log

复制代码

-- 查看当前连接数
SHOW STATUS LIKE 'Threads_connected';
SHOW STATUS LIKE 'Max_used_connections';

-- 查看连接数限制
SHOW VARIABLES LIKE 'max_connections';

-- 查看所有连接
SHOW PROCESSLIST;
SHOW FULL PROCESSLIST;

-- 查看连接来源统计
SELECT host, COUNT(*) FROM information_schema.processlist 
GROUP BY host ORDER BY COUNT(*) DESC;

-- 查看连接状态分布
SELECT command, COUNT(*) FROM information_schema.processlist 
GROUP BY command;

复制代码

# 临时增加连接数
mysql> SET GLOBAL max_connections = 1000;

# 永久配置
vim /etc/my.cnf
[mysqld]
max_connections = 1000

# 查找空闲连接
mysql> SELECT * FROM information_schema.processlist 
WHERE command = 'Sleep' AND time > 60;

# 批量终止空闲连接
mysql> SELECT CONCAT('KILL ', id, ';') FROM information_schema.processlist 
WHERE command = 'Sleep' AND time > 3600;

复制代码

-- 查看超时设置
SHOW VARIABLES LIKE '%timeout%';

-- 常见超时参数
-- connect_timeout: 连接超时
-- wait_timeout: 非交互式连接超时
-- interactive_timeout: 交互式连接超时

-- 调整超时设置
SET GLOBAL wait_timeout = 600;
SET GLOBAL connect_timeout = 30;

复制代码

-- 1. 检查当前连接数
mysqladmin -u root -p status

-- 2. 查看最大连接数
SHOW VARIABLES LIKE 'max_connections';

-- 3. 紧急处理（使用预留连接）
-- 预留连接需要root或具有CONNECTION_ADMIN权限的用户
mysql -u root -p -h localhost --protocol=socket

-- 4. 终止大量空闲连接
SELECT CONCAT('KILL ', id, ';') FROM information_schema.processlist 
WHERE command = 'Sleep' AND time > 300 INTO OUTFILE '/tmp/kill.sql';
SOURCE /tmp/kill.sql;

-- 5. 临时增加连接数
SET GLOBAL max_connections = 2000;

复制代码

# 检查绑定地址
netstat -tlnp | grep 3306

# 检查bind-address配置
mysql> SHOW VARIABLES LIKE 'bind_address';

# 检查用户权限
mysql> SELECT user, host FROM mysql.user WHERE user='appuser';

# 添加远程访问权限
mysql> GRANT ALL PRIVILEGES ON *.* TO 'appuser'@'%' IDENTIFIED BY 'password';
mysql> FLUSH PRIVILEGES;

# 检查防火墙
iptables -L -n | grep 3306
firewall-cmd --list-all

复制代码

#!/bin/bash
# connection_check.sh - 连接问题排查脚本

echo "=== MySQL连接问题排查 ==="
echo "时间: $(date)"
echo "=========================="

# 1. 检查MySQL进程
echo "1. MySQL进程状态:"
ps aux | grep mysqld | grep -v grep

# 2. 检查端口监听
echo -e "\n2. 端口监听状态:"
netstat -tlnp | grep 3306

# 3. 检查最大连接数配置
echo -e "\n3. 连接数配置:"
mysql -u root -p -e "SHOW VARIABLES LIKE 'max_connections';"

# 4. 查看当前连接数
echo -e "\n4. 当前连接状态:"
mysql -u root -p -e "SHOW STATUS LIKE 'Threads_connected';"
mysql -u root -p -e "SHOW STATUS LIKE 'Max_used_connections';"

# 5. 查看连接来源
echo -e "\n5. 连接来源统计:"
mysql -u root -p -e "SELECT host, COUNT(*) FROM information_schema.processlist GROUP BY host ORDER BY COUNT(*) DESC;"

# 6. 查看空闲连接
echo -e "\n6. 空闲连接统计:"
mysql -u root -p -e "SELECT COUNT(*) FROM information_schema.processlist WHERE command='Sleep' AND time > 60;"

# 7. 检查错误日志
echo -e "\n7. 最近10条错误日志:"
tail -10 /var/log/mysql/error.log 2>/dev/null || tail -10 /var/lib/mysql/*.err 2>/dev/null

echo -e "\n=== 排查完成 ==="

复制代码

-- 1. 查看当前执行的查询
SHOW PROCESSLIST;

-- 2. 查看慢查询数量
SHOW STATUS LIKE 'Slow_queries';

-- 3. 查看查询统计
SELECT * FROM performance_schema.events_statements_summary_by_digest 
ORDER BY SUM_TIMER_WAIT DESC LIMIT 10;

-- 4. 查看当前线程状态
SELECT * FROM performance_schema.threads WHERE PROCESSLIST_STATE IS NOT NULL;

复制代码

# 使用top查看MySQL进程CPU使用
top -p $(pgrep mysqld)

# 使用perf分析
perf top -p $(pgrep mysqld)

# 使用strace跟踪
strace -p $(pgrep mysqld) -c

复制代码

-- 1. 开启慢查询日志
SET GLOBAL slow_query_log = ON;
SET GLOBAL long_query_time = 2;
SET GLOBAL log_queries_not_using_indexes = ON;

-- 2. 查看慢查询配置
SHOW VARIABLES LIKE 'slow%';
SHOW VARIABLES LIKE 'long_query_time';

-- 3. 查看慢查询日志
-- Linux: /var/log/mysql/slow.log

-- 4. 使用pt-query-digest分析
pt-query-digest /var/log/mysql/slow.log > slow_report.txt

-- 5. 临时分析慢查询
SELECT * FROM information_schema.processlist WHERE command='Query' AND time > 5;

复制代码

-- 1. 查看当前锁信息
SHOW ENGINE INNODB STATUS\G

-- 2. 查看锁等待
SELECT * FROM information_schema.INNODB_LOCKS;
SELECT * FROM information_schema.INNODB_LOCK_WAITS;

-- 3. 查看当前事务
SELECT * FROM information_schema.INNODB_TRX;

-- 4. 查看阻塞事务
SELECT 
    waiting_trx_id,
    waiting_thread,
    blocking_trx_id,
    blocking_thread
FROM sys.innodb_lock_waits;

-- 5. 终止阻塞事务
KILL <thread_id>;

复制代码

# 1. 查看磁盘I/O
iostat -x 1 10

# 2. 查看MySQL I/O统计
mysql> SHOW STATUS LIKE 'Innodb_data_reads';
mysql> SHOW STATUS LIKE 'Innodb_data_writes';
mysql> SHOW STATUS LIKE 'Innodb_buffer_pool_reads';

# 3. 查看文件打开情况
lsof | grep mysql | wc -l
mysql> SHOW VARIABLES LIKE 'open_files_limit';

# 4. 查看表缓存
mysql> SHOW STATUS LIKE 'Open_tables';
mysql> SHOW STATUS LIKE 'Opened_tables';

复制代码

-- 1. 查看内存相关配置
SHOW VARIABLES LIKE '%buffer%';
SHOW VARIABLES LIKE '%cache%';

-- 2. 查看InnoDB缓冲池使用
SHOW STATUS LIKE 'Innodb_buffer_pool_pages%';

-- 3. 计算缓冲池命中率
SELECT 
    (SELECT variable_value FROM performance_schema.global_status 
     WHERE variable_name='Innodb_buffer_pool_read_requests') / 
    (SELECT variable_value FROM performance_schema.global_status 
     WHERE variable_name='Innodb_buffer_pool_reads') * 100 AS hit_ratio;

-- 4. 查看临时表使用
SHOW STATUS LIKE 'Created_tmp%';

复制代码

# 查看MySQL内存使用
ps aux | grep mysqld
pmap -x $(pgrep mysqld) | head -20

# 查看系统内存
free -h
vmstat 1

复制代码

-- 1. 检查表
CHECK TABLE table_name;

-- 2. 修复表
REPAIR TABLE table_name;

-- 3. 强制修复
REPAIR TABLE table_name USE_FRM;

-- 4. 使用mysqlcheck
mysqlcheck -u root -p --check --all-databases
mysqlcheck -u root -p --repair --all-databases

-- 5. 查看错误信息
SHOW ENGINE INNODB STATUS\G

复制代码

-- 1. 检查主从数据一致性
-- 使用pt-table-checksum
pt-table-checksum --host=master --user=root --password=pass

-- 2. 修复数据不一致
pt-table-sync --sync-to-master --host=slave --user=root --password=pass

-- 3. 校验数据库
mysqlcheck -u root -p --check --all-databases --extended

-- 4. 分析表
ANALYZE TABLE table_name;
OPTIMIZE TABLE table_name;

复制代码

-- 1. 查看死锁信息
SHOW ENGINE INNODB STATUS\G
-- 在输出中查找 "LATEST DETECTED DEADLOCK"

-- 2. 启用死锁日志
SET GLOBAL innodb_print_all_deadlocks = ON;

-- 3. 查询当前锁等待
SELECT 
    r.trx_id AS waiting_trx_id,
    r.trx_mysql_thread_id AS waiting_thread,
    r.trx_query AS waiting_query,
    b.trx_id AS blocking_trx_id,
    b.trx_mysql_thread_id AS blocking_thread,
    b.trx_query AS blocking_query
FROM information_schema.innodb_lock_waits w
JOIN information_schema.innodb_trx r ON w.requesting_trx_id = r.trx_id
JOIN information_schema.innodb_trx b ON w.blocking_trx_id = b.trx_id;

-- 4. 终止阻塞事务
KILL <blocking_thread_id>;

复制代码

-- 1. 查看数据库大小
SELECT 
    table_schema AS 'Database',
    ROUND(SUM(data_length + index_length) / 1024 / 1024, 2) AS 'Size (MB)'
FROM information_schema.tables
GROUP BY table_schema
ORDER BY SUM(data_length + index_length) DESC;

-- 2. 查看表大小
SELECT 
    table_name,
    ROUND(data_length / 1024 / 1024, 2) AS 'Data (MB)',
    ROUND(index_length / 1024 / 1024, 2) AS 'Index (MB)'
FROM information_schema.tables
WHERE table_schema = 'database_name'
ORDER BY data_length DESC;

-- 3. 清理碎片
OPTIMIZE TABLE table_name;

-- 4. 回收未使用空间
ALTER TABLE table_name ENGINE=InnoDB;

复制代码

# 1. 从备份恢复
mysql -u root -p database_name < backup.sql

# 2. 从二进制日志恢复
mysqlbinlog mysql-bin.000001 --start-datetime="2024-01-01 10:00:00" --stop-datetime="2024-01-01 11:00:00" | mysql -u root -p

# 3. 从特定位置恢复
mysqlbinlog mysql-bin.000001 --start-position=123 --stop-position=456 | mysql -u root -p

# 4. 恢复误删除的表
# 从备份中提取表
sed -n '/CREATE TABLE `table_name`/,/UNLOCK TABLES/p' backup.sql > table.sql
mysql -u root -p database_name < table.sql

复制代码

-- 1. 查看复制状态
SHOW SLAVE STATUS\G

-- 重点关注：
-- Slave_IO_Running: Yes/No
-- Slave_SQL_Running: Yes/No
-- Last_IO_Error
-- Last_SQL_Error
-- Seconds_Behind_Master

-- 2. 常见错误及处理

-- 错误1: 1062 - Duplicate entry
STOP SLAVE;
SET GLOBAL SQL_SLAVE_SKIP_COUNTER = 1;
START SLAVE;

-- 错误2: 1032 - Can't find record
STOP SLAVE;
SET GLOBAL SQL_SLAVE_SKIP_COUNTER = 1;
START SLAVE;

-- 错误3: 1452 - Cannot add or update a child row
-- 手动插入缺失数据
INSERT INTO child_table (id) VALUES (xxx);
START SLAVE;

复制代码

-- 1. 查看延迟
SHOW SLAVE STATUS\G
-- 查看 Seconds_Behind_Master

-- 2. 查看中继日志
SHOW SLAVE STATUS\G
-- 查看 Relay_Log_File 和 Relay_Log_Pos

-- 3. 优化并行复制
SET GLOBAL slave_parallel_workers = 4;
SET GLOBAL slave_parallel_type = 'LOGICAL_CLOCK';

-- 4. 查看复制线程
SHOW PROCESSLIST;

复制代码

#!/bin/bash
# replication_check.sh - 复制故障排查脚本

echo "=== MySQL复制状态检查 ==="
echo "时间: $(date)"
echo "=========================="

# 获取复制状态
SLAVE_STATUS=$(mysql -u root -p -e "SHOW SLAVE STATUS\G")

# 检查I/O线程
IO_RUNNING=$(echo "$SLAVE_STATUS" | grep "Slave_IO_Running:" | awk '{print $2}')
SQL_RUNNING=$(echo "$SLAVE_STATUS" | grep "Slave_SQL_Running:" | awk '{print $2}')
SECONDS_BEHIND=$(echo "$SLAVE_STATUS" | grep "Seconds_Behind_Master:" | awk '{print $2}')

echo "I/O线程状态: $IO_RUNNING"
echo "SQL线程状态: $SQL_RUNNING"
echo "复制延迟: $SECONDS_BEHIND 秒"

# 检查错误
LAST_IO_ERROR=$(echo "$SLAVE_STATUS" | grep "Last_IO_Error:" | cut -d':' -f2-)
LAST_SQL_ERROR=$(echo "$SLAVE_STATUS" | grep "Last_SQL_Error:" | cut -d':' -f2-)

if [ -n "$LAST_IO_ERROR" ] && [ "$LAST_IO_ERROR" != " " ]; then
    echo -e "\nI/O错误: $LAST_IO_ERROR"
fi

if [ -n "$LAST_SQL_ERROR" ] && [ "$LAST_SQL_ERROR" != " " ]; then
    echo -e "\nSQL错误: $LAST_SQL_ERROR"
fi

# 检查主库连接
MASTER_HOST=$(echo "$SLAVE_STATUS" | grep "Master_Host:" | awk '{print $2}')
echo -e "\n主库连接: $MASTER_HOST"
ping -c 2 $MASTER_HOST > /dev/null 2>&1
if [ $? -eq 0 ]; then
    echo "主库网络可达"
else
    echo "主库网络不可达"
fi

# 检查中继日志
RELAY_LOG=$(echo "$SLAVE_STATUS" | grep "Relay_Log_File:" | awk '{print $2}')
echo "当前中继日志: $RELAY_LOG"

echo -e "\n=== 检查完成 ==="

复制代码

# 1. CPU
# 选择高频CPU，多核心

# 2. 内存
# InnoDB缓冲池建议设置为物理内存的70-80%
# 计算公式：内存 = 系统内存 - (OS内存 + 其他进程内存)

# 3. 磁盘
# 使用SSD，RAID10
# 分离数据目录和日志目录

# 4. 网络
# 使用万兆网络
# 分离复制网络和应用网络

复制代码

# /etc/sysctl.conf 内核参数优化
# 网络优化
net.ipv4.tcp_tw_reuse = 1
net.ipv4.tcp_fin_timeout = 30
net.ipv4.tcp_max_tw_buckets = 5000
net.ipv4.tcp_max_syn_backlog = 8192
net.ipv4.tcp_keepalive_time = 600
net.core.somaxconn = 65535

# 文件系统
fs.file-max = 6553500
fs.nr_open = 6553500

# 内存
vm.swappiness = 10
vm.dirty_ratio = 30
vm.dirty_background_ratio = 5

# 应用参数
sysctl -p

# 系统限制
cat >> /etc/security/limits.conf << EOF
mysql soft nofile 65535
mysql hard nofile 65535
mysql soft nproc 65535
mysql hard nproc 65535
EOF

复制代码

# /etc/my.cnf 完整优化配置

[client]
port = 3306
socket = /var/lib/mysql/mysql.sock
default-character-set = utf8mb4

[mysql]
prompt="\\u@\\h [\\d]> "
default-character-set = utf8mb4
no-auto-rehash

[mysqld]
# 基础设置
user = mysql
port = 3306
socket = /var/lib/mysql/mysql.sock
pid-file = /var/run/mysqld/mysqld.pid
basedir = /usr
datadir = /var/lib/mysql
tmpdir = /tmp

# 字符集
character-set-server = utf8mb4
collation-server = utf8mb4_unicode_ci
init-connect = 'SET NAMES utf8mb4'
skip-character-set-client-handshake

# 连接设置
max_connections = 1000
max_connect_errors = 100
connect_timeout = 10
wait_timeout = 600
interactive_timeout = 28800
max_allowed_packet = 64M

# 线程设置
thread_cache_size = 256
thread_stack = 256K

# 缓冲区设置
key_buffer_size = 256M
sort_buffer_size = 2M
read_buffer_size = 2M
read_rnd_buffer_size = 4M
join_buffer_size = 2M

# 临时表
tmp_table_size = 64M
max_heap_table_size = 64M

# 表缓存
table_open_cache = 2048
table_definition_cache = 2048

# InnoDB核心设置
default-storage-engine = InnoDB
innodb_buffer_pool_size = 8G                 # 物理内存的70-80%
innodb_buffer_pool_instances = 8
innodb_buffer_pool_chunk_size = 128M
innodb_log_file_size = 2G
innodb_log_buffer_size = 16M
innodb_flush_log_at_trx_commit = 2
innodb_flush_method = O_DIRECT
innodb_file_per_table = 1
innodb_open_files = 2048
innodb_io_capacity = 2000
innodb_io_capacity_max = 4000
innodb_read_io_threads = 4
innodb_write_io_threads = 4
innodb_purge_threads = 4
innodb_page_cleaners = 4
innodb_adaptive_hash_index = ON
innodb_lock_wait_timeout = 50

# 事务设置
transaction_isolation = READ-COMMITTED
autocommit = 1

# 日志设置
log_error = /var/log/mysql/error.log
slow_query_log = 1
slow_query_log_file = /var/log/mysql/slow.log
long_query_time = 2
log_queries_not_using_indexes = 1
min_examined_row_limit = 100

# 二进制日志
server-id = 1
log_bin = /var/lib/mysql/mysql-bin
binlog_format = ROW
binlog_row_image = full
expire_logs_days = 7
max_binlog_size = 100M
sync_binlog = 1
binlog_cache_size = 32K

# GTID（如果需要）
gtid_mode = OFF
enforce_gtid_consistency = OFF

# 复制设置
relay_log = /var/lib/mysql/mysql-relay-bin
relay_log_recovery = 1
slave_parallel_workers = 4
slave_parallel_type = LOGICAL_CLOCK

# 性能优化
performance_schema = ON
performance_schema_consumer_events_statements_history_long = ON
performance_schema_consumer_events_statements_history = ON

# 安全设置
local_infile = 0
skip_symbolic_links = yes
secure_file_priv = /var/lib/mysql-files

[mysqldump]
quick
quote-names
max_allowed_packet = 64M

[mysqld_safe]
log-error = /var/log/mysql/error.log
pid-file = /var/run/mysqld/mysqld.pid

复制代码

-- 1. 查看缓冲池大小
SHOW VARIABLES LIKE 'innodb_buffer_pool_size';

-- 2. 查看缓冲池使用情况
SHOW STATUS LIKE 'Innodb_buffer_pool_pages%';

-- 3. 查看缓冲池命中率
SELECT 
    (SELECT variable_value FROM performance_schema.global_status 
     WHERE variable_name='Innodb_buffer_pool_read_requests') /
    (SELECT variable_value FROM performance_schema.global_status 
     WHERE variable_name='Innodb_buffer_pool_reads') * 100 AS hit_ratio;

-- 4. 设置缓冲池大小
SET GLOBAL innodb_buffer_pool_size = 8589934592;  -- 8GB

-- 5. 设置缓冲池实例数
SET GLOBAL innodb_buffer_pool_instances = 8;

复制代码

-- 1. 查看优化器开关
SHOW VARIABLES LIKE 'optimizer_switch';

-- 2. 启用优化器特性
SET GLOBAL optimizer_switch = 'index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,engine_condition_pushdown=on,index_condition_pushdown=on,mrr=on,mrr_cost_based=on,block_nested_loop=on,batched_key_access=off,materialization=on,semijoin=on,loosescan=on,firstmatch=on,duplicateweedout=on,subquery_materialization_cost_based=on,use_index_extensions=on,condition_fanout_filter=on,derived_merge=on';

-- 3. 查看统计信息
ANALYZE TABLE table_name;

-- 4. 查看表统计信息
SELECT * FROM information_schema.statistics WHERE table_schema='database_name' AND table_name='table_name';

复制代码

-- 1. 选择合适的数据类型
-- 使用INT代替VARCHAR存储数字
-- 使用ENUM代替VARCHAR存储有限选项
-- 使用TIMESTAMP代替DATETIME

-- 2. 优化索引
-- 为WHERE、ORDER BY、GROUP BY字段创建索引
-- 避免在索引列上使用函数
-- 使用覆盖索引

-- 3. 分区表
CREATE TABLE orders (
    id INT NOT NULL,
    order_date DATE NOT NULL,
    amount DECIMAL(10,2)
) PARTITION BY RANGE (YEAR(order_date)) (
    PARTITION p2022 VALUES LESS THAN (2023),
    PARTITION p2023 VALUES LESS THAN (2024),
    PARTITION p2024 VALUES LESS THAN (2025)
);

-- 4. 查看表碎片
SELECT 
    table_name,
    ROUND(data_free / 1024 / 1024, 2) AS '碎片(MB)'
FROM information_schema.tables
WHERE table_schema = 'database_name'
ORDER BY data_free DESC;

-- 5. 整理碎片
OPTIMIZE TABLE table_name;

复制代码

-- 1. 避免SELECT *
-- 不好的写法
SELECT * FROM users WHERE id = 1;

-- 好的写法
SELECT id, name, email FROM users WHERE id = 1;

-- 2. 使用EXISTS代替IN（当子查询数据量大时）
-- 不好的写法
SELECT * FROM users WHERE id IN (SELECT user_id FROM orders);

-- 好的写法
SELECT * FROM users u WHERE EXISTS (SELECT 1 FROM orders o WHERE o.user_id = u.id);

-- 3. 使用JOIN代替子查询
-- 不好的写法
SELECT u.name, (SELECT COUNT(*) FROM orders WHERE user_id = u.id) AS order_count FROM users u;

-- 好的写法
SELECT u.name, COUNT(o.id) AS order_count 
FROM users u 
LEFT JOIN orders o ON u.id = o.user_id 
GROUP BY u.id;

-- 4. 使用UNION ALL代替UNION
-- UNION会去重，UNION ALL不会

-- 5. 使用批量操作
-- 不好的写法
INSERT INTO users (name) VALUES ('user1');
INSERT INTO users (name) VALUES ('user2');

-- 好的写法
INSERT INTO users (name) VALUES ('user1'), ('user2');

-- 6. 分页优化
-- 不好的写法
SELECT * FROM orders LIMIT 100000, 20;

-- 好的写法
SELECT * FROM orders WHERE id > 100000 LIMIT 20;

复制代码

-- 1. 查看索引使用情况
SELECT 
    index_name,
    COUNT(*) as usage_count
FROM performance_schema.table_io_waits_summary_by_index_usage
WHERE object_schema = 'database_name'
GROUP BY index_name
ORDER BY usage_count;

-- 2. 查找未使用的索引
SELECT 
    table_name,
    index_name
FROM information_schema.statistics
WHERE table_schema = 'database_name'
AND index_name NOT IN (
    SELECT index_name 
    FROM performance_schema.table_io_waits_summary_by_index_usage
    WHERE object_schema = 'database_name'
);

-- 3. 删除重复索引
-- 使用pt-duplicate-key-checker
pt-duplicate-key-checker --host=localhost --user=root --password=pass

-- 4. 索引选择原则
-- 高选择性字段优先
-- 联合索引遵循最左前缀原则
-- 避免在频繁更新的列上建索引

复制代码

#!/bin/bash
# backup_optimize.sh - 优化备份脚本

# 配置
BACKUP_DIR="/backup/mysql"
DATE=$(date +%Y%m%d_%H%M%S)
MYSQL_USER="root"
MYSQL_PASS="password"

# 1. 使用压缩备份
mysqldump -u$MYSQL_USER -p$MYSQL_PASS --all-databases \
    --single-transaction \
    --quick \
    --master-data=2 \
    --routines \
    --triggers \
    --events | gzip > $BACKUP_DIR/full_$DATE.sql.gz

# 2. 分库备份
for db in $(mysql -u$MYSQL_USER -p$MYSQL_PASS -e "SHOW DATABASES;" | grep -Ev "Database|information_schema|performance_schema|mysql|sys"); do
    mysqldump -u$MYSQL_USER -p$MYSQL_PASS --databases $db \
        --single-transaction \
        --quick | gzip > $BACKUP_DIR/${db}_$DATE.sql.gz
done

# 3. 使用xtrabackup热备份
innobackupex --user=$MYSQL_USER --password=$MYSQL_PASS \
    --stream=tar ./ | gzip > $BACKUP_DIR/xtrabackup_$DATE.tar.gz

# 4. 备份二进制日志
mysqlbinlog --read-from-remote-server \
    --host=localhost \
    --user=$MYSQL_USER \
    --password=$MYSQL_PASS \
    --raw \
    --result-file=$BACKUP_DIR/binlog_$DATE_ \
    mysql-bin.000001

# 5. 清理过期备份
find $BACKUP_DIR -type f -mtime +7 -delete

echo "备份完成: $BACKUP_DIR/full_$DATE.sql.gz"

复制代码

#!/bin/bash
# mysql_performance_monitor.sh - 性能监控脚本

# 配置
MYSQL_USER="root"
MYSQL_PASS="password"
LOG_FILE="/var/log/mysql/performance.log"
ALERT_EMAIL="admin@example.com"

# 监控函数
monitor_performance() {
    local timestamp=$(date '+%Y-%m-%d %H:%M:%S')
    
    # 获取关键指标
    local qps=$(mysql -u$MYSQL_USER -p$MYSQL_PASS -e "SHOW STATUS LIKE 'Queries';" | grep Queries | awk '{print $2}')
    local connections=$(mysql -u$MYSQL_USER -p$MYSQL_PASS -e "SHOW STATUS LIKE 'Threads_connected';" | grep Threads_connected | awk '{print $2}')
    local slow_queries=$(mysql -u$MYSQL_USER -p$MYSQL_PASS -e "SHOW STATUS LIKE 'Slow_queries';" | grep Slow_queries | awk '{print $2}')
    local buffer_hit=$(mysql -u$MYSQL_USER -p$MYSQL_PASS -e "SELECT ROUND((SELECT variable_value FROM performance_schema.global_status WHERE variable_name='Innodb_buffer_pool_read_requests') / (SELECT variable_value FROM performance_schema.global_status WHERE variable_name='Innodb_buffer_pool_reads') * 100, 2) AS hit_ratio;" | grep -v hit_ratio)
    
    # 写入日志
    echo "$timestamp | QPS: $qps | Connections: $connections | Slow: $slow_queries | BufferHit: $buffer_hit%" >> $LOG_FILE
    
    # 告警检查
    if [ $connections -gt 800 ]; then
        echo "连接数过高: $connections" | mail -s "MySQL Alert: High Connections" $ALERT_EMAIL
    fi
    
    if [ $slow_queries -gt 100 ]; then
        echo "慢查询过多: $slow_queries" | mail -s "MySQL Alert: High Slow Queries" $ALERT_EMAIL
    fi
}

# 主循环
while true; do
    monitor_performance
    sleep 60
done

复制代码

#!/bin/bash
# mysql_tuner.sh - MySQL自动优化脚本

# 配置
MYSQL_USER="root"
MYSQL_PASS="password"
CONFIG_FILE="/etc/my.cnf"

# 收集当前配置
get_config() {
    local var=$1
    mysql -u$MYSQL_USER -p$MYSQL_PASS -e "SHOW VARIABLES LIKE '$var';" | grep $var | awk '{print $2}'
}

# 获取系统内存
TOTAL_MEM=$(free -g | grep Mem | awk '{print $2}')
INNODB_BUFFER=$(($TOTAL_MEM * 70 / 100))

# 生成优化建议
cat > /tmp/mysql_optimize.sh << EOF
#!/bin/bash
# 自动生成的优化脚本

echo "开始MySQL优化..."

# 备份当前配置
cp $CONFIG_FILE ${CONFIG_FILE}.backup

# 应用优化配置
cat >> $CONFIG_FILE << EOC

# ========== 自动优化配置 ==========
# 内存: ${TOTAL_MEM}GB
innodb_buffer_pool_size = ${INNODB_BUFFER}G

# 连接设置
max_connections = 1000
thread_cache_size = 256

# 缓存设置
key_buffer_size = 256M
tmp_table_size = 64M
max_heap_table_size = 64M

# InnoDB设置
innodb_log_file_size = 2G
innodb_flush_log_at_trx_commit = 2
innodb_flush_method = O_DIRECT

# 日志设置
slow_query_log = 1
long_query_time = 2

# ========== 优化配置结束 ==========
EOC

echo "配置已更新，请重启MySQL"
echo "systemctl restart mysqld"
EOF

chmod +x /tmp/mysql_optimize.sh
echo "优化脚本已生成: /tmp/mysql_optimize.sh"

复制代码

□ 连接问题检查
  □ 网络连接
  □ 连接数配置
  □ 用户权限
  □ 防火墙设置

□ 性能问题检查
  □ CPU使用率
  □ 内存使用
  □ 磁盘I/O
  □ 慢查询日志

□ 复制问题检查
  □ 复制状态
  □ 网络连接
  □ 数据一致性
  □ 延迟时间

□ 数据问题检查
  □ 表完整性
  □ 索引有效性
  □ 锁等待
  □ 死锁情况

MySQL故障排查与生产环境优化

一、MySQL故障排查概述

1. 故障排查方法论

2. 故障排查工具清单

3. 关键日志文件

二、连接类故障排查

1. 连接数过高

2. 连接超时

3. Too many connections 错误

4. 连接被拒绝

5. 连接排查脚本

三、性能类故障排查

1. CPU使用率过高

2. 慢查询分析

3. 锁等待分析

4. I/O瓶颈分析

5. 内存使用分析

四、数据类故障排查

1. 表损坏修复

2. 数据不一致

3. 死锁分析

4. 表空间不足

5. 数据恢复

五、复制类故障排查

1. 复制中断

2. 复制延迟

3. 复制故障排查脚本

六、生产环境优化

1. 硬件层面优化

2. 操作系统优化

3. MySQL参数优化

4. InnoDB缓冲池优化

5. 查询优化器优化

6. 表结构优化

7. SQL优化示例

8. 索引优化建议

9. 备份策略优化

10. 性能监控脚本

11. 自动优化脚本

七、总结

故障排查清单

优化最佳实践