> -----Original Message-----
> From: Eric Gunnett [mailto:[EMAIL PROTECTED] 
> Sent: Tuesday, January 11, 2005 9:36 AM
> To: mysql@lists.mysql.com
> Subject: Tuning MySQL
> 
>       I have a quad processor server, with 4 gigs of memory. 
> It is only running MySQL right now and seems really slow. Can 
> someone give me a few suggestions on optimizing My.cnf file 
> for this system.  We are running 
> mysql-standard-4.0.23-pc-linux-i68, on it. Here is the my.cnf file

1st how many disks are in the server. What is your key cache hit
percentage? This can be figured out by 

Issue SHOW STATUS

Take 

Key_reads/Key_read_requests * 100 give the %


Next look at handler_read_rnd_next

If this number is high then your doing table scans so optimizing your
keys would be better.

Additional comments below.


> 
> # The MySQL server
> [mysqld]
> port            = 3306
> socket          = /tmp/mysql.sock
> skip-locking
> key_buffer = 384M

You will prob have to raise the key buffer

> max_allowed_packet = 2M
> table_cache = 512
> sort_buffer_size = 2M
> read_buffer_size = 2M
> myisam_sort_buffer_size = 64M

Add tmp_table_size=16M

> thread_cache = 8

Look at your connection rate if it's high raise the above

> query_cache_size = 32M
> # Try number of CPU's*2 for thread_concurrency thread_concurrency = 8

Thread_concurrency is only a valid option for solaris

> 
> user=mysql
> basedir=/usr/local/mysql
> datadir=/usr2/mysql/data
> max_connections=400
> max_connect_errors=300
> interactive_timeout=2400
> wait_timeout=60
> back_log=100
> #skip-networking
> server-id       = 2
> 
> [isamchk]
> key_buffer = 256M
> sort_buffer_size = 256M
> read_buffer = 2M
> write_buffer = 2M
> 
> [myisamchk]
> key_buffer = 256M
> sort_buffer_size = 256M
> read_buffer = 2M
> write_buffer = 2M
> 
> 
> 
> --
> MySQL General Mailing List
> For list archives: http://lists.mysql.com/mysql
> To unsubscribe:    
> http://lists.mysql.com/[EMAIL PROTECTED]
> 
> 

--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:    http://lists.mysql.com/[EMAIL PROTECTED]

Reply via email to