15 messages in com.mysql.lists.win32Re: mysqldumpslow| From | Sent On | Attachments |
|---|---|---|
| Joelle Tegwen | 14 Apr 2006 12:01 | |
| jbon...@sola.com.au | 17 Apr 2006 16:30 | |
| sandhya | 18 Apr 2006 02:51 | |
| Daniel da Veiga | 18 Apr 2006 09:45 | |
| Shawn Green | 18 Apr 2006 10:23 | |
| Na Derro Cartwright | 18 Apr 2006 12:22 | |
| jbon...@sola.com.au | 18 Apr 2006 17:33 | |
| Armando | 18 Apr 2006 21:31 | |
| Joelle Tegwen | 18 May 2006 08:16 | |
| Joelle Tegwen | 18 May 2006 10:38 | |
| Tobin Cataldo | 18 May 2006 11:31 | |
| Joelle Tegwen | 18 May 2006 13:30 | |
| Tobin Cataldo | 19 May 2006 06:18 | |
| Joelle Tegwen | 19 May 2006 08:39 | |
| Joelle Tegwen | 22 May 2006 07:07 |
| Subject: | Re: mysqldumpslow |
|---|---|
| From: | Tobin Cataldo (tcat...@bham.lib.al.us) |
| Date: | 05/19/2006 06:18:04 AM |
| List: | com.mysql.lists.win32 |
To get the lines from the MSDOS prompt try -> right-click select all, ctrl-c. Then paste in TextPad or something.
I have read that mysqldumpslow doesn't handle prepared statements (not sure if this is applicable)...
You might try running the mysqldumpslow on a limited test. In fact, grab the examples below and run the mysqldumpslow against it, and compare it against the results I got (shown below). If you get the same, then I would suggest looking at the structure of your slowlog... Or change slowlog to a new log temporarily (mysql server needs to be restarted making a change to the logs) run a few queries in the mysql> prompt (this way you know they are not 'prepared' queries), then check against mysqldumpslow.
My slowlog looks like this: C:\mysql\bin\mysqld-max-nt, Version: 5.0.18-nt-max-log. started with: TCP Port: 3306, Named Pipe: (null) Time Id Command Argument # Time: 060518 13:28:33 # User@Host: root[root] @ localhost [127.0.0.1] # Query_time: 0 Lock_time: 0 Rows_sent: 1 Rows_examined: 1 SHOW VARIABLES LIKE 'datadir'; # User@Host: root[root] @ localhost [127.0.0.1] # Query_time: 0 Lock_time: 0 Rows_sent: 1 Rows_examined: 1 SHOW VARIABLES LIKE 'log_error'; C:\mysql\bin\mysqld-max-nt, Version: 5.0.18-nt-max-log. started with: TCP Port: 3306, Named Pipe: (null) Time Id Command Argument C:\mysql\bin\mysqld-max-nt, Version: 5.0.18-nt-max-log. started with: TCP Port: 3306, Named Pipe: (null) Time Id Command Argument
My output looks like this: C:\mysql\scripts>perl mysqldumpslow -s=t -t=2 c:\mysql\data\slowlog2.txt
Reading mysql slow query log from c:\mysql\data\slowlog2.txt Count: 1 Time=0.00s (0s) Lock=0.00s (0s) Rows=0.0 (0), 0users@0hosts C:\mysql\bin\mysqld-max-nt, Version: N.N.N-nt-max-log. started with: TCP Port: N, Named Pipe: (null) # Time: N N:N:N # User@Host: root[root] @ localhost [N.N.N.N] # Query_time: N Lock_time: N Rows_sent: N Rows_examined: N SHOW VARIABLES LIKE 'S'
Count: 1 Time=0.00s (0s) Lock=0.00s (0s) Rows=1.0 (1), root[root]@localhost SHOW VARIABLES LIKE 'S'; C:\mysql\bin\mysqld-max-nt, Version: N.N.N-nt-max-log. started with: TCP Port: N, Named Pipe: (null) C:\mysql\bin\mysqld-max-nt, Version: N.N.N-nt-max-log. started with: TCP Port: N, Named Pipe: (null)
C:\mysql\scripts>
Tobin Cataldo
Joelle Tegwen wrote:
Ok, so I didn't get the error, but I get this huge long scroll of SQL statements with a space between each letter. I can't even see the beginning of the scroll because it maxes out the buffer.
I thought that saying -t=1 would only give me one query am I misunderstanding this?
Thanks much Joelle
Tobin Cataldo wrote:
I was able to reproduce the error by setting -t= <Number greater than actual queries in log>
Try setting -t = 1 or -t = 2. Or take a look at the log txt file and see if you have at least 5 queries in there.
Tobin Cataldo
Joelle Tegwen wrote:
ok, so I got it running... thanks..
and I type C:\Program Files\MySQL\MySQL Server 5.0\scripts>perl mysqldumpslow -s=t -t=5 c:\mysql\slowlog20060518.txt and I get this big long scroll of sql statements and then this: Died at mysqldumpslow line 137, <> chunk 1.
What does that mean? Thanks Joelle