MySQL 5.1.61 caused high disk IO wait

[Robert W.]

Hello there,


We just upgraded from Centos 5.4 to Centos 6.2 and MySQL 5.0.77 to 5.1.61. Using mysqldump, we exported and re-imported all database file. Everything works smoothly.

However, the websites are noticably slow on the same hardware and Internet connection. After poking around for a long while, I noticed very high disk IO wait %, in 30~45%, which is definitely caused by MySQL. I'm very puzzled by the IO wait increase.

tuning-primer.sh points the finger at temp table on disk being 37%, see below.

TEMP TABLES
Current max_heap_table_size = 200 M
Current tmp_table_size = 200 M
Of 29069 temp tables, 37% were created on disk
Perhaps you should increase your tmp_table_size and/or max_heap_table_size
to reduce the number of disk-based temporary tables
Note! BLOB and TEXT columns are not allow in memory tables.
If you are using these columns raising these values might not impact your
ratio of on disk temp tables.

I have doubled max_heap_table_size and tmp_table_size from 100M and the on-disk temp table remains at 37%.

No configuration and code change to the website was made during the transition. Apache configuration remian the same.

Can anybody shed some light on the difference between MySQL 5.1.61 and 5.0.77? Thanks in advance.

Robert