Sunday, 15 March 2015

MySQL configuration: memory usage "dangerously high"? -



MySQL configuration: memory usage "dangerously high"? -

i'm running centos directadmin on dedicated server (quad core i7 ~8gb ram) , i've encountered major issues mysql, on more details can found here.

i've run perl mysql tuner script this website, seek , determine problem is.

sadly, output doesn't create lot of sense me:

>> mysqltuner 1.2.0 - major hayden <major@mhtx.net> >> bug reports, feature requests, , downloads @ http://mysqltuner.com/ >> run '--help' additional options , output filtering please come in mysql administrative login: root please come in mysql administrative password: -------- general statistics -------------------------------------------------- [--] skipped version check mysqltuner script [ok] running supported mysql version 5.5.20-log [ok] operating on 64-bit architecture -------- storage engine statistics ------------------------------------------- [--] status: +archive -bdb -federated +innodb -isam -ndbcluster [--] info in myisam tables: 236m (tables: 686) [--] info in innodb tables: 226m (tables: 199) [--] info in performance_schema tables: 0b (tables: 17) [!!] total fragmented tables: 312 -------- security recommendations ------------------------------------------- [ok] database users have passwords assigned -------- performance metrics ------------------------------------------------- [--] for: 18h 42m 34s (1m q [18.265 qps], 49k conn, tx: 3b, rx: 835m) [--] reads / writes: 60% / 40% [--] total buffers: 528.0m global + 7.0m per thread (1500 max threads) [!!] maximum possible memory usage: 10.8g (138% of installed ram) [ok] slow queries: 0% (3/1m) [ok] highest usage of available connections: 2% (37/1500) [ok] key buffer size / total myisam indexes: 64.0m/74.1m [ok] key buffer nail rate: 99.9% (8m cached / 5k reads) [ok] query cache efficiency: 66.9% (529k cached / 792k selects) [!!] query cache prunes per day: 83513 [ok] sorts requiring temporary tables: 0% (0 temp sorts / 42k sorts) [!!] joins performed without indexes: 5962 [ok] temporary tables created on disk: 5% (5k on disk / 97k total) [ok] thread cache nail rate: 99% (37 created / 49k connections) [ok] table cache nail rate: 42% (512 open / 1k opened) [ok] open file limit used: 10% (756/7k) [ok] table locks acquired immediately: 99% (513k immediate / 513k locks) [!!] innodb info size / buffer pool: 226.7m/128.0m -------- recommendations ----------------------------------------------------- general recommendations: run optimize table defragment tables improve performance mysql started within lastly 24 hours - recommendations may inaccurate cut down overall mysql memory footprint scheme stability adjust bring together queries utilize indexes variables adjust: *** mysql's maximum memory usage dangerously high *** *** add together ram before increasing mysql buffer variables *** query_cache_size (> 64m) join_buffer_size (> 2.0m, or utilize indexes joins) innodb_buffer_pool_size (>= 226m)

is tuner suggesting mysql installation requires more 8gb ram?

the main website on server gets ~30k unique visitors/week, then, certainly mysql isn't beingness hammered that hard?

the settings in my.cnf are:

[mysqld] local-infile=0 bind-address=127.0.0.1 max_connections=1500 long_query_time=2 log-slow-queries=/var/log/mysql_slow.log expire_logs_days=14 external-locking open_files_limit=1536 key_buffer_size=64m innodb_buffer_pool_size=128m innodb_additional_mem_pool_size=8m innodb_log_buffer_size=8m group_concat_max_len=16k max_sort_length=16k max_length_for_sort_data=16k query_cache_type=1 query_cache_limit=4m query_cache_size=64m innodb_thread_concurrency=16 thread_concurrency=16 thread_cache=128 thread_stack=1m read_buffer_size=1m join_buffer_size=2m read_rnd_buffer_size=1m table_cache=512 tmp_table_size=256m max_heap_table_size=256m log-warnings [mysqld_safe] log-error=/var/log/mysqld.log

does have advice how can optimise mysql installation?

assuming utilize innodb looks on first glance not using ram db humming. how this:

innodb_buffer_pool_size = 4g innodb_log_file_size = 1g

mysql

No comments:

Post a Comment