I have a dedicated server running Mariadb 11.1.3. I am trying to tune the database using mysqltuner.pl. In the results I got this results on the Innodb:
-------- InnoDB Metrics ----------------------------------------------------------------------------
[--] InnoDB is enabled.
[OK] InnoDB File per table is activated
[OK] InnoDB Buffer Pool size ( 1.0G ) under limit for 64 bits architecture: (17179869184.0G )
[OK] InnoDB buffer pool / data size: 1.0G / 493.0M
[OK] Ratio InnoDB log file size / InnoDB Buffer pool size: 256.0M * 1/1.0G should be equal to 25%
[--] Number of InnoDB Buffer Pool Chunk: 64 for 1 Buffer Pool Instance(s)
[OK] Innodb_buffer_pool_size aligned with Innodb_buffer_pool_chunk_size & Innodb_buffer_pool_instances
[!!] InnoDB Read buffer efficiency: 0% (-19040 hits / 0 total)
[!!] InnoDB Write Log efficiency: 88.69% (239467 hits / 270003 total)
[OK] InnoDB log waits: 0.00% (0 waits / 30536 writes)
I searched the internet for the reason why the read buffer efficiency is 0% but I can't find anything about it.
My my.cnf:
[client-server]
socket = /run/mysqld/mysqld.sock
[mysqld]
skip_name_resolve = ON
join_buffer_size = 1M
tmp_table_size = 64M
max_heap_table_size = 64M
table_definition_cache = 4096
performance_schema = ON
innodb_buffer_pool_size = 1024M
query_cache_limit=8M
query_cache_size=256M
query_cache_type=1
innodb_log_file_size = 256M
slow_query_log = 1
slow-query_log_file = /var/log/mysql-slow.log
long_query_time = 1
!includedir /etc/mysql/conf.d/
!includedir /etc/mysql/mariadb.conf.d/
Any ideas where to look for this problem?
It was a bug in the 11.1.3 release, Innodb_buffer_pool_read_requests wasn't being incremented hence negative hits. It will be fixed in the next release 11.1.4.