**导读**
> 作者：杨漆
> 16年关系型数据库管理，从oracle 9i 、10g、11g、12c到Mysql5.5、5.6、5.7、8.0 到TiDB获得3个OCP、2个OCM；运维路上不平坦，跌过不少坑、熬过许多夜。把工作笔记整理出来分享给大伙儿，希望帮到大家少走弯路、少熬夜。
使用Mysql中如果CPU在95%及以上，Qps突然增到2万以上，这时Mysql随时有死去风险。
这时该怎么办？
应急方法：
第一: 先限制Innodb的并发处理.如果innodb_thread_concurrency = 0 可以先改成 16或64 看机器压力,如果 非常大,先改成16让机器的压力下来,然后慢慢增大,适应自已的业务.
set global innodb_thread_concurrency=16;
第二: 对于连接数已经超过600的情况,可以适当的限制一下连接数,宁可让前端报一下错,也别让DB挂. 只要DB活着总是可以用来加载一下数据,慢慢的DB压力也会降下来的. 限制单用户连接数在300以下
set global max_user_connections=300;
关闭 innodb_stats_on_metadata防止对读取information_schema时造成大量读取磁盘进行信息统计 （有些监控程序从这里抓取数据，会终止）
set global innodb_stats_on_metadata=0;
================================
应急处置完毕后进行RootCause分析：
思路：
1、确定高负载的类型 htop，dstat命令看负载高是CPU还是IO
2、监控具体的sql语句，是insert update 还是 delete导致高负载
3、检查mysql慢日志
打开慢查询方法：vi my.cnf,在[mysqld]如下几行:
log_slow_queries = /data/slow.log #慢查询日志路径
long_query_time = 1 #记录SQL查询超过1s的语句
log-queries-not-using-indexes = 1 #记录没有使用索引的sql
4、检查硬件问题
dstat
看具体哪个用户哪个进程占用了相关系统资源，当前CPU、内存谁在使用
1. # dstat -l -m -r -c --top-io --top-mem --top-cpu
2. --io/total- ------memory-usage----- --most-expensive- ----most-expensive---- -most-expensive-
3. read writ| used buff cach free| memory process | i/o process | cpu process
4. 1.90 267 |3399M 178M 3892M 400M|php-fpm: poo 372M|init 1682k 647k|flush-202:0 0.1
5. 0 72.0 |3399M 178M 3892M 400M|php-fpm: poo 372M|php-fpm: po 10k 143k|php-fpm: pool2.0
6. 0 8.00 |3399M 178M 3892M 399M|php-fpm: poo 372M|nginx: work 228k 229k|php-fpm: pool0.5
7. 0 88.0 |3399M 178M 3892M 399M|php-fpm: poo 372M|nginx: work 102k 166k|php-fpm: pool 11
8. 0 38.0 |3399M 178M 3892M 399M|php-fpm: poo 372M|php-fpm: po 787k 650B|php-fpm: pool4.8
9. 0 0 |3399M 178M 3892M 399M|php-fpm: poo 372M|php-fpm: po 788k 723B|php-fpm: pool1.8
10. 0 140 |3400M 178M 3892M 399M|php-fpm: poo 372M|nginx: work 38k 154k|php-fpm: pool1.2
11. 0 12.0 |3400M 178M 3892M 399M|php-fpm: poo 372M|nginx: work 178k 364k|php-fpm: pool1.5
12. 0 0 |3400M 178M 3892M 399M|php-fpm: poo 372M|nginx: work 758k 639k|php-fpm: pool1.5
13. 0 12.0 |3400M 178M 3892M 399M|php-fpm: poo 372M|nginx: work 773k 616k|php-fpm: pool2.0
14. 6.00 0 |3401M 178M 3892M 398M|php-fpm: poo 372M|nginx: work 994k 688k|nginx: worker1.5
15. 0 272 |3401M 178M 3892M 398M|php-fpm: poo 372M|nginx: work 388k 422k|php-fpm: pool1.5
16. 0 0 |3400M 178M 3893M 398M|php-fpm: poo 372M|nginx: work 483k 548k|php-fpm: pool1.8
17. 0 4.00 |3400M 178M 3893M 398M|php-fpm: poo 372M|php-fpm: po 787k 650B|php-fpm: pool1.5
18. 0 12.0 |3400M 178M 3893M 398M|php-fpm: poo 372M|nginx: work 223k 323k|php-fpm: pool1.5
19. 0 0 |3400M 178M 3893M 398M|php-fpm: poo 372M|nginx: work 371k 474k|php-fpm: pool7.8
Htop ( htop是top的增强版，更直观)
1. # htop
2. 1 [||||||||||| 12.4%]
3. 2 [||||||||| 9.5%]
4. 3 [| 1.0%]
5. 4 [|| 1.9%]
6. Mem[|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||3394/7869MB]
7. Swp[|||||||||||||| 75/478MB]
8. Tasks: 71, 12 thr; 2 running
9. Load average: 0.39 0.39 0.31
10. Uptime: 319 days(!), 19:01:17
11. PID USER PRI NI VIRT RES SHR S CPU% MEM% TIME Command
12. 1 root 20 0 19232 396 248 S 0.0 0.0 0:01.86 /sbin/init
13. 30752 root 20 0 52532 72 56 S 0.0 0.0 0:00.16 ├─ /usr/sbin/vsftpd /etc/vsftpd/vsftpd.conf
14. 24301 root 20 0 193M 3268 1600 S 0.0 0.0 1:41.43 ├─ /usr/sbin/snmpd -LS0-6d -Lf /dev/null -p /var/run/snmpd.pid
15. 21361 root 20 0 902M 6500 1308 S 0.0 0.1 0:07.16 ├─ php-fpm: master process (/etc/php-fpm.conf)
16. 28627 www 20 0 962M 202M 138M S 0.0 2.6 0:34.46 │ ├─ php-fpm: pool www-c
17. 27537 www 20 0 965M 236M 171M R 1.4 3.0 1:19.64 │ ├─ php-fpm: pool www-c
18. 27449 www 20 0 961M 251M 189M S 0.0 3.2 1:35.54 │ ├─ php-fpm: pool www-a
19. 26442 www 20 0 962M 280M 217M S 0.0 3.6 2:29.71 │ ├─ php-fpm: pool www-a
20. 26310 www 20 0 917M 251M 234M S 1.9 3.2 2:46.45 │ ├─ php-fpm: pool www-a
21. 26162 www 20 0 962M 297M 233M S 0.0 3.8 2:37.50 │ ├─ php-fpm: pool www-b
22. 26147 www 20 0 924M 258M 233M S 0.0 3.3 2:38.37 │ ├─ php-fpm: pool www-c
23. 25717 www 20 0 965M 302M 238M S 0.0 3.8 2:54.50 │ ├─ php-fpm: pool www-c
24. 24585 www 20 0 964M 324M 260M S 0.0 4.1 4:15.20 │ ├─ php-fpm: pool www-b
tcpdump
抓取mysql包分析，一般抓3306端口的数据
[root@cc ~]# tcpdump -i eth0 -A -s 3000 port 3306 > ~/sql.log
然后使用awk,sort,wc 等命令进行分析
或 用 # tcpdump -i eth0 -s 0 -l -w - dst port 3306 | strings | perl -e 'while(<>) { chomp; next if /^[^ ] [ ]*$/;
if(/^(SELECT|UPDATE|DELETE|INSERT|SET|COMMIT|ROLLBACK|CREATE|DROP|ALTER)/i) {if (defined $q) { print "$qn"; }$q=$_;} else {$_ =~ s/^[ t] //; $q.=" $_";}}'
就可以看出最繁忙的sql语句
strace 或 pstack 查看系统调用是否有问题，进程是否堵塞，是否有Broken pipe
[root@cc ~]# strace -p 26578
[root@cc ~]# pstack 26356
分析mysql慢日志，查看哪些sql语句最耗时
# pt-query-digest slow.logs
1. # 390ms USER TIME, 10ms system TIME, 15.67M rss, 105.84M vsz
2. # CURRENT DATE: Thu DEC 29 13:22:42 2014
3. # Hostname: test
4. # Files: slow.log
5. # Overall: 776 total, 11 UNIQUE, 0.00 QPS, 0.00x concurrency _____________
6. # TIME range: 2021-07-30 04:03:19 TO 2021-07-3005:02:51
7. # Attribute total MIN MAX avg 95% stddev median
8. # ============ ======= ======= ======= ======= =======
9. # EXEC TIME 5657s 2s 33s 7s 23s 6s 5s
10. # LOCK TIME 33s 0 19s 43ms 98us 715ms 38us
11. # ROWS sent 323.38k 0 107.36k 426.73 0.99 6.35k 0
12. # ROWS examine 323.39k 0 107.36k 426.74 0 6.35k 0
13. # Query SIZE 217.95k 38 562 287.61 420.77 81.78 284.79
查看系统到底在干什么
mysql> show full processlist;
1. ----------- --------------- ---------------------
2. | Id | User | Host | db | Command | Time | State | Info |
3. ----------- --------------- ---------------------
4. | 184498848 | testdb_rr1356 | 10.11.211.120:61343 | testdb_rr1356_db121 | Sleep | 1384 | | NULL |
5. | 184508740 | testdb_rr1356 | 10.11.211.120:11809 | testdb_rr1356_db121 | Sleep | 87 | | NULL |
6. | 184509415 | testdb_rr1356 | 10.11.211.120:12760 | testdb_rr1356_db121 | Query | 0 | NULL | show full processlist |
7. | 184509451 | testdb_rr1356 | 10.11.211.120:12804 | testdb_rr1356_db121 | Sleep | 10 | | NULL |
8. | 184509528 | testdb_rr1356 | 10.11.211.120:12919 | testdb_rr1356_db121 | Query | 0 | freeing items | DESCRIBE test_channel |
检查mysql配置参数是否有问题，引起大量的IO或者高CPU操作
innodb_flush_log_at_trx_commit、innodb_buffer_pool_size 、key_buffer_size 等重要参数
1. mysql> show variables like '%innodb%';
2. ----------------------------- ----------------------------
3. | Variable_name | Value |
4. ----------------------------- ----------------------------
5. | have_innodb | YES |
6. | ignore_builtin_innodb | ON |
7. | innodb_adaptive_flushing | ON |
8. | innodb_adaptive_hash_index | ON |
9. | innodb_additional_mem_pool_size | 2097152 |
10. | innodb_autoextend_increment | 8 |
11. | innodb_autoinc_lock_mode | 1 |
12. | innodb_buffer_pool_size | 2013265920 |
13. | innodb_change_buffering | inserts |
14. | innodb_checksums | ON |
查看当前事务，内存使用情况
1. mysql> show engine innodb status G
2. LATEST DETECTED DEADLOCK
3. ------------------------
4. 150731 10:36:50
5. *** (1) TRANSACTION:
6. TRANSACTION EBFBBEC, ACTIVE 0 sec, process no 20691, OS thread id 47345217033984 inserting
7. mysql tables in use 1, locked 1
8. LOCK WAIT 5 lock struct(s), heap size 1248, 4 row lock(s), undo log entries 2
9. MySQL thread id 143249904, query id 1286731854 10.135.21.120 tybuser2014 update
10. BUFFER POOL AND MEMORY
11. ----------------------
12. Total memory allocated 2058485760; in additional pool allocated 0
13. Dictionary memory allocated 819282
14. Buffer pool size 122879
15. Free buffers 97599
16. Database pages 24313
17. Old database pages 8954
18. Modified db pages 7
19. Pending reads 0
20. Pending writes: LRU 0, flush list 0, single page 0
21. Pages made young 6, not young 0
22. 0.00 youngs/s, 0.00 non-youngs/s
23. Pages read 1049, created 41853, written 30401604
24. 0.00 reads/s, 0.00 creates/s, 1.75 writes/s
25. Buffer pool hit rate 1000 / 1000, young-making rate 0 / 1000 not 0 / 1000
26. Pages read ahead 0.00/s, evicted without access 0.00/s, Random read ahead 0.00/s
27. LRU len: 24313, unzip_LRU len: 0
28. I/O sum[45]:cur[0], unzip sum[0]:cur[0]
最后通过zabbix或者cacti等监控来查看IO、CPU、MEMORY、磁盘等是否有异常
通过以上基本就可以把问题找出来了
