query_cache_size=0 is not enough

Last week at the OUG Harmony conference thanks to Dimitri Kravtchuk I learned that setting query_cache_size=0 does not disable and remove locking from the Query Cache. You actually need to also set query_cache_type=0. This appears to been a bug, seen in the presently still open MySQL bugs database entry #38511 .

My recommendation to customers now is to set both variables on all existing MySQL versions if you are not using the MySQL Query Cache.

Thanks to the Performance Schema in MySQL 5.5 for uncovering this. More information in Dimitri’s detailed post at MySQL Performance: Using Performance Schema

Details of all MySQL presentations at OUG Harmony 2011 in Helsinki, Finland.

Tagged with: Databases MySQL

Related Posts

MySQL and Heatwave Summit Presentation

MySQL and Heatwave Summit Presentation

Last week I had the opportunity to speak at the MySQL and Heatwave Summit in San Francisco. I discussed the impact of the new MySQL 8.0 default caching_sha2_password authentication, replacing the mysql_native_password authentication that was the default for approximately 20 of the 30 years that MySQL has existed.

Read more
Readyset QueryPilot Announcement

Readyset QueryPilot Announcement

At the MySQL and Heatwave Summit 2025 today, Readyset announced a new data systems architecture pattern named Readyset QueryPilot . This architecture which can front a MySQL or PostgreSQL database infrastructure, combines the enterprise-grade ProxySQL and Readyset caching with intelligent query monitoring and routing to help support applications scale and produce more predictable results with varied workloads.

Read more
More CPUs or Newer CPUs

More CPUs or Newer CPUs

In a CPU-bound database workload, regardless of price, would you scale-up or scale-new? What if price was the driving factor, would you scale-up or scale-new? I am using as a baseline the first available AWS Graviton2 processor for RDS (r6g).

Read more