pt-query-digest解析MySQL Binlog日志文件

pt-query-digest解析MySQL Binlog日志文件

解析binlog工具百家争鸣，最常用的是mysqlbinlog，各有千秋，对于DBA，唯手熟尔罢了
然而，有工具的地方就有江湖，故本文无意争论工具属优属劣，免得引起不成熟的争端

仅介绍pt-query-digest其实也是能用来解析Binlog，友好、可读性强、便于快速诊断故障

如果直接：
[root@ld88 mysqldata]# pt-query-digest --type binlog mysql-bin88.000189
将无法解析，需先做如下转换
[root@ld88 mysqldata]# mysqlbinlog mysql-bin88.000189 > mysql-bin88.000189.sql

看两个例子，以下为了排版，对输出做了大量裁减-_-!

㈠ 指定时间窗口

[root@ld88 mysqldata]# pt-query-digest --type binlog --since "2013-11-06 20:55:00" --until "2013-11-06 21:00:00" mysql-bin88.000189.sql
mysql-bin88.000189.sql:  17% 02:22 remain
mysql-bin88.000189.sql:  33% 01:58 remain
mysql-bin88.000189.sql:  47% 01:39 remain
mysql-bin88.000189.sql:  62% 01:12 remain
mysql-bin88.000189.sql:  83% 00:28 remain

# 160.9s user time, 8.8s system time, 23.62M rss, 150.04M vsz
# Current date: Thu Nov  7 15:37:19 2013
# Hostname: ld88
# Files: mysql-bin88.000189.sql
# Overall: 914 total, 31 unique, 3.06 QPS, 41.19kx concurrency ___________
# Time range: 2013-11-06 20:55:00 to 20:59:59
# Attribute          total    min    max    avg    95%  stddev  median
# ============    ======= ======= ======= ======= ======= ======= =======
# Exec time        12316866s  13349s  13639s  13476s  13443s    145s  13443s
# Query size      442.35k      6  35.33k  434.29 1012.63  1.88k  143.84
# error code            0      0      0      0      0      0      0

# Profile
# Rank Query ID          Response time      Calls R/Call    V/M  Item
# ==== ================== ================== ===== ========== ===== ======
#    1 0x972882477A1D4A3F 3739234.0000 30.4%  277 13499.0397  1.54 UPDATE tbBlogArticleStat?
#    2 0xF5B3ADEC45DB5266 1099848.0000  8.9%    82 13412.7805  1.39 UPDATE tbBlogArticleStat?
#    3 0xC05BF1F3A8344559 1099848.0000  8.9%    82 13412.7805  1.39 UPDATE tbBlogArticleChart
#    4 0xA85CE0CC3154666E 1024921.0000  8.3%    76 13485.8026  1.26 UPDATE tbBlogTag
#    5 0x7C12B8C66B369B73  822705.0000  6.7%    61 13486.9672  1.26 INSERT tbBlogTagArticle
#    6 0xE8059EB28F9F68AA  752337.0000  6.1%    56 13434.5893  1.39 INSERT tbBlogArticleVote
#    7 0x01FCF322381E8B7E  404606.0000  3.3%    30 13486.8667  1.28 INSERT tbBlogArticleDayClick
#    8 0xD5C00F71DB944F6C  404606.0000  3.3%    30 13486.8667  1.28 UPDATE tbBlogSubArticleStatist
#    9 0x6EFA3A2E4EC6B202  404606.0000  3.3%    30 13486.8667  1.28 UPDATE tbBlogMemberSort
#  10 0xE14AB2C787449950  404606.0000  3.3%    30 13486.8667  1.28 UPDATE tbBlogMemberStat
#  11 0x34764E44BE970CDD  404606.0000  3.3%    30 13486.8667  1.28 INSERT tbBlogArticleChart
#  12 0xCFEB2F244234CE05  404605.0000  3.3%    30 13486.8333  1.28 INSERT tbBlogArticleStat?
#  13 0xCDB381C90AF965D4  404604.0000  3.3%    30 13486.8000  1.28 INSERT tbBlogArticle?
#  14 0xF7D29C9021590977  162011.0000  1.3%    12 13500.9167  1.30 UPDATE tbBlogArticleChart
#  15 0xFBA7BD32B7694172  134777.0000  1.1%    10 13477.7000  1.23 INSERT tbBlogTag
#  16 0xB2317C2DE1E87251  108184.0000  0.9%    8 13523.0000  0.00 UPDATE tbBlogMemberStat
# MISC 0xMISC              540762.0000  4.4%    40 13519.0500  0.0 <14 ITEMS>

# Query 1: 0.93 QPS, 12.59kx concurrency, ID 0x972882477A1D4A3F at byte 218115626
# This item is included in the report because it matches --limit.
# Scores: V/M = 1.54
# Time range: 2013-11-06 20:55:00 to 20:59:57
# Attribute    pct  total    min    max    avg    95%  stddev  median
# ============ === ======= ======= ======= ======= ======= ======= =======
# Count        30    277
# Exec time    30 3739234s  13351s  13639s  13499s  13443s    144s  13443s
# Query size    10  44.59k    144    201  164.84  192.76  17.53  166.51
# error code    0      0      0      0      0      0      0      0
# Query_time distribution
#  1us
#  10us
# 100us
#  1ms
#  10ms
# 100ms
#    1s
#  10s+  ################################################################
# Tables
#    SHOW TABLE STATUS LIKE 'tbBlogArticleStat0022'\G
#    SHOW CREATE TABLE `tbBlogArticleStat0022`\G
update tbBlogArticleStat0022 set PrevArticleID='0',PrevArticleAppearTime='',NextArticleID='35810718',NextArticleAppearTime='2011-03-14 19:59:01' where ArticleID='35605800'\G
# Converted for EXPLAIN
# EXPLAIN /*!50100 PARTITIONS*/
select  PrevArticleID='0',PrevArticleAppearTime='',NextArticleID='35810718',NextArticleAppearTime='2011-03-14 19:59:01' from tbBlogArticleStat0022 where  ArticleID='35605800'\G

更多详情见请继续阅读下一页的精彩内容：

如何安全删除MySQL下的binlog日志

MySQL--binlog日志恢复数据

MySQL删除binlog日志及日志恢复数据的方法

MySQL binlog三种格式介绍及分析

MySQL 利用binlog增量备份+还原实例

MySQL删除binlog日志及日志恢复数据

MySQL binlog预分配的实现和性能