MYSQL 慢查询日志分析

mysqlsla的安装与使用

1.下载 mysqlsla
[root@localhost tmp]# wget http://hackmysql.com/scripts/mysqlsla-2.03.tar.gz --19:45:45-- http://hackmysql.com/scripts/mysqlsla-2.03.tar.gz Resolving hackmysql.com... 64.13.232.157
Connecting to hackmysql.com|64.13.232.157|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 33674 (33K) [application/x-tar]
Saving to: `mysqlsla-2.03.tar.gz.2'

100%[===========================================================================================>] 33,674 50.2K/s in 0.7s

19:45:47 (50.2 KB/s) - `mysqlsla-2.03.tar.gz.2' saved [33674/33674]

2.解压
[root@localhost tmp]# tar -zxvf mysqlsla-2.03.tar.gz
mysqlsla-2.03/
mysqlsla-2.03/Changes
mysqlsla-2.03/INSTALL
mysqlsla-2.03/README
mysqlsla-2.03/Makefile.PL
mysqlsla-2.03/bin/
mysqlsla-2.03/bin/mysqlsla
mysqlsla-2.03/META.yml
mysqlsla-2.03/lib/
mysqlsla-2.03/lib/mysqlsla.pm
mysqlsla-2.03/MANIFEST
[root@localhost tmp]# cd mysqlsla-2.03
[root@localhost mysqlsla-2.03]# ls
bin Changes INSTALL lib Makefile.PL MANIFEST META.yml README

3.执行perl脚本检查包依赖关系
[root@localhost mysqlsla-2.03]# perl Makefile.PL
Checking if your kit is complete...
Looks good
Writing Makefile for mysqlsla

4.安装
[root@localhost mysqlsla-2.03]# make && make install;
cp lib/mysqlsla.pm blib/lib/mysqlsla.pm
cp bin/mysqlsla blib/script/mysqlsla
/usr/bin/perl "-MExtUtils::MY" -e "MY->fixin(shift)" blib/script/mysqlsla
Manifying blib/man3/mysqlsla.3pm
Installing /usr/lib/perl5/site_perl/5.8.8/mysqlsla.pm
Installing /usr/share/man/man3/mysqlsla.3pm
Installing /usr/bin/mysqlsla
Writing /usr/lib/perl5/site_perl/5.8.8/i386-linux-thread-multi/auto/mysqlsla/.packlist
Appending installation info to /usr/lib/perl5/5.8.8/i386-linux-thread-multi/perllocal.pod

5.简单使用

语法：
Slow log: mysqlsla -lt slow slow.log
General log: mysqlsla -lt general general.log
Binary log: mysqlbinlog bin.log | mysqlsla -lt binary -

这里以slow log为例：

[root@localhost mysqlsla-2.03]# mysqlsla -lt slow /tmp/127_slow.log | more
Report for slow logs: /tmp/127_slow.log
24 queries total, 6 unique
Sorted by 't_sum'
Grand Totals: Time 16 s, Lock 1 s, Rows sent 18, Rows Examined 2.10M

______________________________________________________________________ 001 ___
Count : 18 (75.00%)
Time : 15 s total, 833.333 ms avg, 0 to 8 s max (93.75%)
95% of Time : 7 s total, 411.765 ms avg, 0 to 4 s max
Lock Time (s) : 0 total, 0 avg, 0 to 0 max (0.00%)
95% of Lock : 0 total, 0 avg, 0 to 0 max
Rows sent : 0 avg, 0 to 0 max (0.00%)
Rows examined : 116.51k avg, 8 to 1.05M max (99.99%)
Database :
Users :
root@localhost : 100.00% (18) of query, 100.00% (24) of all users

Query abstract:
INSERT INTO t2 SELECT * FROM t2;

Query sample:
insert into t2 select * from t2;
........

选项说明：
总查询次数 (queries total)， 去重后的sql数量 (unique)
输出报表的内容排序(sorted by)
最重大的慢sql统计信息, 包括 平均执行时间, 等待锁时间, 结果行的总数, 扫描的行总数.
Count, sql的执行次数及占总的slow log数量的百分比.
Time, 执行时间, 包括总时间, 平均时间, 最小, 最大时间, 时间占到总慢sql时间的百分比.
95% of Time, 去除最快和最慢的sql, 覆盖率占95%的sql的执行时间.
Lock Time, 等待锁的时间.
95% of Lock , 95%的慢sql等待锁时间.
Rows sent, 结果行统计数量, 包括平均, 最小, 最大数量.
Rows examined, 扫描的行数量.
Database, 属于哪个数据库
Users, 哪个用户,IP, 占到所有用户执行的sql百分比
Query abstract, 抽象后的sql语句
Query sample, sql语句

mysqlsla常用参数说明：

1) -log-type (-lt) type logs:
通过这个参数来制定log的类型，主要有slow, general, binary, msl, udl,分析slow log时通过制定为slow.

2) -sort:
制定使用什么参数来对分析结果进行排序，默认是按照t_sum来进行排序。
t_sum:按总时间排序
c_sum:按总次数排序
c_sum_p: sql语句执行次数占总执行次数的百分比。

3) -top:
显示sql的数量，默认是10,表示按规则取排序的前多少条

4) –statement-filter (-sf) [+-][TYPE]:
过滤sql语句的类型，比如select、update、drop.
[TYPE]有SELECT, CREATE, DROP, UPDATE, INSERT，例如"+SELECT,INSERT"，不出现的默认是-，即不包括。

5) db：要处理哪个库的日志：

例如，只取backup库的select语句、按c_sum_p排序的前2条记录

[root@localhost mysqlsla-2.03]# mysqlsla -lt slow -sort c_sum_p -sf "+select" -db backup -top 2 /tmp/127_slow.log
Report for slow logs: /tmp/127_slow.log
4 queries total, 3 unique
Sorted by 'c_sum_p'
Grand Totals: Time 1 s, Lock 1 s, Rows sent 18, Rows Examined 195

______________________________________________________________________ 001 ___
Count : 2 (50.00%)
Time : 0 total, 0 avg, 0 to 0 max (0.00%)
Lock Time (s) : 0 total, 0 avg, 0 to 0 max (0.00%)
Rows sent : 1 avg, 1 to 1 max (11.11%)
Rows examined : 86 avg, 77 to 94 max (87.69%)
Database :
Users :
root@localhost : 100.00% (2) of query, 100.00% (4) of all users

Query abstract:
SELECT SUM(format(duration,N)) AS duration FROM information_schema.profiling WHERE query_id=N;

Query sample:
select sum(format(duration,6)) as duration from information_schema.profiling where query_id=7;

______________________________________________________________________ 002 ___
Count : 1 (25.00%)
Time : 1 s total, 1 s avg, 1 s to 1 s max (100.00%)
Lock Time (s) : 1 s total, 1 s avg, 1 s to 1 s max (100.00%)
Rows sent : 4 avg, 4 to 4 max (22.22%)
Rows examined : 12 avg, 12 to 12 max (6.15%)
Database :
Users :
root@localhost : 100.00% (1) of query, 100.00% (4) of all users

Query abstract:
SELECT * FROM tt WHERE a=N;

Query sample:
select * from tt where a=2;
本文出自 “ohgenlong16300blog.com” 博客，请务必保留此出处http://ohgenlong16300.blog.51cto.com/499130/1272993