我正在通过hikari-cp Clojure库在Clojure应用程序中使用HikariCP。
该应用程序为多个都有自己的MySQL数据库和数据库用户的客户端提供服务。每个客户端都有自己的HikariCP连接池,据我了解,这是必需的,因为每个客户端都有专用的数据库用户。添加新客户端后,我开始从MySQL收到间歇性的“太多连接”错误。我猜每个客户端的默认池大小为10个连接会累加太多的连接(超过了MySQL中max_connections的默认设置151)。
每个客户端的数据库总负载不是很大。
简单地减少每个客户端的连接数似乎很麻烦,因为每当添加新客户端时,都必须减小池的大小。
将max_connections设置为与客户端数量成正比的数字(例如50 + [客户端数量] * 10)是否安全?
还是有一种方法可以使用相同的池连接到不同的数据库(具有不同的数据库用户)?
答案 0 :(得分:0)
Here is an example测试文件不使用连接池。您可以为单个命令,一组命令或每个事务创建新的数据库连接:
(def raw-db-spec
{:classname "org.h2.Driver"
:subprotocol "h2:mem" ; the prefix `jdbc:` is added automatically
:subname "demo;DB_CLOSE_DELAY=-1" ; `;DB_CLOSE_DELAY=-1` very important!!!
; http://www.h2database.com/html/features.html#in_memory_databases
; http://makble.com/using-h2-in-memory-database-in-clojure
:user "sa" ; "system admin"
:password "" ; empty string by default
})
(dotest
; creates & drops a connection (& transaction) for each command
(jdbc/db-do-commands raw-db-spec ["drop table if exists langs"
"drop table if exists releases"])
; Creates and uses a connection for all commands
(jdbc/with-db-connection
[conn raw-db-spec]
(jdbc/db-do-commands
conn
[(jdbc/create-table-ddl :langs
[[:id :serial]
[:lang "varchar not null"]])
(jdbc/create-table-ddl :releases
[[:id :serial]
[:desc "varchar not null"]
[:langId "numeric"]])]))
; create & use a connection for multiple commands
(jdbc/with-db-connection
[conn raw-db-spec]
(jdbc/insert-multi! raw-db-spec :langs ; => ({:id 1} {:id 2})
[{:lang "Clojure"}
{:lang "Java"}])
(let [result (jdbc/query raw-db-spec ["select * from langs"])]
(is= result [{:id 1, :lang "Clojure"}
{:id 2, :lang "Java"}])))
; Wraps all commands in a single transaction
(jdbc/with-db-transaction
[tx raw-db-spec]
(let [clj-id (grab :id (only (jdbc/query tx ["select id from langs where lang='Clojure'"])))]
(jdbc/insert-multi! tx :releases
[{:desc "ancients" :langId clj-id}
{:desc "1.8" :langId clj-id}
{:desc "1.9" :langId clj-id}]))
(let [java-id (grab :id (only (jdbc/query tx ["select id from langs where lang='Java'"])))]
(jdbc/insert-multi! tx :releases
[{:desc "dusty" :langId java-id}
{:desc "8" :langId java-id}
{:desc "9" :langId java-id}
{:desc "10" :langId java-id}])))
Here is a different file显示了Hikari数据库池的使用。
(def datasource-options-sample {:auto-commit true
:read-only false
:connection-timeout 30000
:validation-timeout 5000
:idle-timeout 600000
:max-lifetime 1800000
:minimum-idle 10
:maximum-pool-size 10
:pool-name "db-pool"
:adapter "h2" ; "postgresql"
:username "sa"
:password ""
:database-name "database"
:server-name "localhost"
:port-number 5432
:register-mbeans false})
(def datasource-options {:adapter "h2"
:url "jdbc:h2:mem:demo;DB_CLOSE_DELAY=-1"
:username "sa"
:password ""})
(def ^:dynamic db-conn nil)
(defn with-connection-pool
"Creates and uses a connection for test function"
[tst-fn]
(let [datasource (pool/make-datasource datasource-options)]
(binding [db-conn {:datasource datasource}]
(tst-fn)
(pool/close-datasource datasource)))) ; close the connection - also closes/destroys the in-memory database
(use-fixtures
:once with-connection-pool) ; use the same db connection pool for all tests
(dotest
; creates & drops a connection (& transaction) for each command
(jdbc/db-do-commands db-conn ["drop table if exists langs"
"drop table if exists releases"])
(jdbc/db-do-commands
db-conn
[(jdbc/create-table-ddl :langs
[[:id :serial]
[:lang "varchar not null"]])
(jdbc/create-table-ddl :releases
[[:id :serial]
[:desc "varchar not null"]
[:langId "numeric"]])])
(jdbc/insert-multi! db-conn :langs ; => ({:id 1} {:id 2})
[{:lang "Clojure"}
{:lang "Java"}])
(let [result (jdbc/query db-conn ["select * from langs"])]
(is= result [{:id 1, :lang "Clojure"}
{:id 2, :lang "Java"}]))
我建议对您的项目进行分叉,并删除所有连接池内容,然后尝试运行它。如果以后要添加,可以使用(jdbc / with-db-connection ...)or
(jdbc / with-db-connection ...)`将多个命令分组为一个数据库连接。