http://www.linuxidc.com/Linux/2012-06/63322.htm
Raw Trace的收集方法
1. 打開Trace，Help > Diagnostics > Trace > Trace > Trace with Binds and Waits
Trace項代表的意思
•No Trace – turns trace off.
•Regular Trace – generates a regular SQL trace by performing the following statement:
ALTER SESSION SET SQL_TRACE = TRUE;
•Trace with Binds – writes bind variable values in the SQL trace file
•Trace with Waits – writes wait events in the SQL trace file
•Trace with Binds and Waits – writes both bind variable values and wait events in the SQL trace file
2.執行業務功能
3.關閉Trace，Help > Diagnostics > Trace > Trace > No Trace
系統會彈出一個窗口，告訴你Trace文件所在的目錄。
這個目錄實際上是數據庫系統參數表(v$parameter)中的user_dump_dest的值，可以執行下邊的SQL來找到Trace文件所在的目錄
1.SELECT value FROM v$parameter WHERE name = 'user_dump_dest';
2.
3.Output:/slot/ems7061/Oracle/db/tech_st/11.2.0/admin/az1mu213_rws60145rems/diag/rdbms/az1mu213/az1mu213/trace
trace文件名後邊的那個數字是Database Server PID，也可以從Help>About中找到。
TKPROF
之前我們收集的trace日誌仍爲Raw Trace，如果要做性能分析的話，往往要轉換爲TKPROF，TKPROF可以把Raw Trace轉換爲更易讀的形式。
1.TKPROF Command
2.$tkprof tracefile outputfile [explain= ] [table= ] [print= ] [insert= ] [sys= ] [sort= ] ...
3.
4.Sample:
5.TKPROF raw_trace.trc OUTPUTA.TKPROF EXPLAIN=scott/tiger SYS=NO SORT=(EXECPU,FCHCPU)
tracefile：你要分析的trace文件
outputfile：格式化後的文件
explain=user/password@connectstring
table=schema.tablename
注1：這兩個參數是一起使用的，通過連接數據庫對在trace文件中出現的每條sql語句查看執行計劃，並將之輸出到outputfile中
注2：該table必須是數據庫中不存在的，如果存在會報錯
print=n：只列出最初N個sql執行語句
insert=filename：會產生一個sql文件，運行此文件可將收集到的數據insert到數據庫表中
sys=no：過濾掉由sys執行的語句
record=filename：可將非嵌套執行的sql語句過濾到指定的文件中去
waits=yes|no：是否統計任何等待事件
aggregate=yes|no：是否將相同sql語句的執行信息合計起來，默認爲yes
sort=option：設置排序選項，選項如下：
prscnt：number of times parse was called
prscpu：cpu time parsing
prsela：elapsed time parsing
prsdsk：number of disk reads during parse
prsqry：number of buffers for consistent read during parse
prscu：number of buffers for current read during parse
prsmis：number of misses in library cache during parse
execnt：number of execute was called
execpu：cpu time spent executing
exeela：elapsed time executing
exedsk：number of disk reads during execute
exeqry：number of buffers for consistent read during execute
execu：number of buffers for current read during execute
exerow：number of rows processed during execute
exemis：number of library cache misses during execute
fchcnt：number of times fetch was called
fchcpu：cpu time spent fetching
fchela：elapsed time fetching
fchdsk：number of disk reads during fetch
fchqry：number of buffers for consistent read during fetch
fchcu：number of buffers for current read during fetch
fchrow：number of rows fetched
userid：userid of user that parsed the cursor
1.Tkprof: Release 9.2.0.1.0 - Production on Tue Dec 24 15:32:43 2002
2.Copyright (c) 1982, 2002, Oracle Corporation. All rights reserved.
3.
4.Trace file: ORCL102_ora_3064.trc
5.
6.Sort options: default
7.
8.********************************************************************************
9.count = number of times OCI procedure was executed
10.cpu = cpu time in seconds executing
11.elapsed = elapsed time in seconds executing
12.disk = number of physical reads of buffers from disk
13.query = number of buffers gotten for consistent read
14.current = number of buffers gotten in current mode (usually for update)
15.rows = number of rows processed by the fetch or execute call
16.********************************************************************************
17.
18.select *
19.from
20. employee where emp_id = 3737
21.
22.call count cpu elapsed disk query current rows
23.------- ------ -------- ---------- ---------- ---------- ---------- ----------
24.Parse 10 0.00 0.03 0 0 0 0
25.Execute 10 0.00 0.00 0 0 0 0
26.Fetch 20 0.34 0.35 72 4730 0 10
27.------- ------ -------- ---------- ---------- ---------- ---------- ----------
28.total 40 0.34 0.39 72 4730 0 10
29.
30.Misses in library cache during parse: 1
31.Optimizer goal: CHOOSE
32.Parsing user id: 59
33.
34.Rows Row Source Operation
35.------- ---------------------------------------------------
36. 1 TABLE ACCESS FULL EMPLOYEE
本篇文章來源於 Linux公社網站(www.linuxidc.com) 原文鏈接：http://www.linuxidc.com/Linux/2012-06/63322p2.htm
query1：Select value from v$parameter
where name = 'user_dump_dest';
query2： Select oracle_process_id
from fnd_concurrent_requests
where request_id = 2754278
command：tkprof dv11_ora_7830_OPERATIONS_CR2759648.trc dv11_ora_7830_OPERATIONS_CR2759648.tkp sort=prsdsk,exedsk,fchdsk
1，在併發程序定義界面勾上trace選項
2，執行併發程序
3，query1找到生成的trc文件的路徑，query2找到文件名中的processing_id
4，tkprof生成tkp文件
5，選取tkp文件中的前5個query在Toad裏Explain Plan
使用tkprof
命令在DV11
要先運行一下環境變量的文件
. ./appenv.sh $env $bash
appenv.sh所在目錄爲：/rdbms/apps1159
要先cd到/u01/oracle/DV11/dv11db/9.2.0/admin/DV11_prcsgisol03/udump目錄下，再執行tkprof tracefile outputfile