Firstly, server is an Athlon x2 4000+ w/ 2GB RAM and a 7200rpm SATA Seagate 250GB hard disk. OS is Debian Etch v4.0. Zabbix server is v1.4.2, MySQL is v5.0, db engine is MyISAM, but Zabbix db is InnoDB.
MySQL reports the number of "key_reads" being around 2000x greater than the number of "key_read_requests" after 24 hours or so. Is this normal? Here's a sampling after the server has been up for around 10 minutes. At this point the key_reads already outnumer key_read_requests. It seems the key_buffer is not being used at all.
| Key_blocks_not_flushed | 0 |
| Key_blocks_unused | 463920 |
| Key_blocks_used | 7 |
| Key_read_requests | 1954 |
| Key_reads | 3663 |
| Key_write_requests | 4271 |
| Key_writes | 9 |
Here are the pertinent values from /etc/mysql/my.cnf:
max_heap_table_size = 128M
tmp_table_size = 96M
key_buffer_size = 128M
innodb_file_per_table
innodb_buffer_pool_size=512M
innodb_log_buffer_size=64M
MySQL reports the number of "key_reads" being around 2000x greater than the number of "key_read_requests" after 24 hours or so. Is this normal? Here's a sampling after the server has been up for around 10 minutes. At this point the key_reads already outnumer key_read_requests. It seems the key_buffer is not being used at all.
| Key_blocks_not_flushed | 0 |
| Key_blocks_unused | 463920 |
| Key_blocks_used | 7 |
| Key_read_requests | 1954 |
| Key_reads | 3663 |
| Key_write_requests | 4271 |
| Key_writes | 9 |
Here are the pertinent values from /etc/mysql/my.cnf:
max_heap_table_size = 128M
tmp_table_size = 96M
key_buffer_size = 128M
innodb_file_per_table
innodb_buffer_pool_size=512M
innodb_log_buffer_size=64M
Comment