一定数量的查询后 HikariCP 连接超时错误

Question

在我开始之前，我想说我已经检查了以下内容，但它们没有帮助我：

https://github.com/brettwooldridge/HikariCP/issues/104

基本上，我得到了一个 HikariCP stracktrace，但我不知道是什么原因造成的。

java.sql.SQLTransientConnectionException: HikariPool-1 - Connection is not available, request timed out after 30000ms.
at com.zaxxer.hikari.pool.HikariPool.createTimeoutException(HikariPool.java:548)
at com.zaxxer.hikari.pool.HikariPool.getConnection(HikariPool.java:186)
at com.zaxxer.hikari.pool.HikariPool.getConnection(HikariPool.java:145)
at com.zaxxer.hikari.HikariDataSource.getConnection(HikariDataSource.java:83)
at de.arraying.Arraybot.managers.ManagerSync.addCustomCommand(ManagerSync.java:192)
at de.arraying.Arraybot.commands.CommandsCommand.onCommand(CommandsCommand.java:100)
at de.arraying.Arraybot.commands.Command.execute(Command.java:72)
at de.arraying.Arraybot.listeners.ListenerChat.onGuildMessageReceived(ListenerChat.java:68)
at net.dv8tion.jda.core.hooks.ListenerAdapter.onEvent(ListenerAdapter.java:299)
at net.dv8tion.jda.core.hooks.InterfacedEventManager.handle(InterfacedEventManager.java:64)
at net.dv8tion.jda.core.handle.MessageCreateHandler.handleDefaultMessage(MessageCreateHandler.java:97)
at net.dv8tion.jda.core.handle.MessageCreateHandler.handleInternally(MessageCreateHandler.java:47)
at net.dv8tion.jda.core.handle.SocketHandler.handle(SocketHandler.java:38)
at net.dv8tion.jda.core.requests.WebSocketClient.handleEvent(WebSocketClient.java:688)
at net.dv8tion.jda.core.requests.WebSocketClient.onTextMessage(WebSocketClient.java:437)
at com.neovisionaries.ws.client.ListenerManager.callOnTextMessage(ListenerManager.java:352)
at com.neovisionaries.ws.client.ReadingThread.callOnTextMessage(ReadingThread.java:262)
at com.neovisionaries.ws.client.ReadingThread.callOnTextMessage(ReadingThread.java:240)
at com.neovisionaries.ws.client.ReadingThread.handleTextFrame(ReadingThread.java:965)
at com.neovisionaries.ws.client.ReadingThread.handleFrame(ReadingThread.java:748)
at com.neovisionaries.ws.client.ReadingThread.main(ReadingThread.java:110)
at com.neovisionaries.ws.client.ReadingThread.run(ReadingThread.java:66)

我试过更改 maximum pool size、minimum idle 并且我还启用了 leak detection（在 2 秒）。 None 其中有帮助，除了我每次执行查询时都会进行泄漏检测，所以可能与此有关。

这是我当前的配置：

    HikariConfig hikariConfig = new HikariConfig();
    hikariConfig.setJdbcUrl("jdbc:mysql://"+url+":3306/"+database+"?useSSL=false");
    hikariConfig.setUsername(username);
    hikariConfig.setPassword(password);
    hikariConfig.setMaximumPoolSize(10);
    hikariConfig.setMinimumIdle(3);
    hikariConfig.setLeakDetectionThreshold(2000);
    dataSource = new HikariDataSource(hikariConfig);

我的查询方法结构如下：

        // inside a try/catch, after some checks that aren't related.
        PreparedStatement preparedStatement =
                dataSource.getConnection().prepareStatement(query);
        preparedStatement.setString(2, id);
        preparedStatement.setString(3, name);
        preparedStatement.setObject(1, value);
        preparedStatement.executeUpdate();
        preparedStatement.close();

我应该在这之后关闭连接吗？我能想到的唯一可能导致错误的是内存泄漏，我认为我没有。我的 CPU 使用也很好，我的互联网连接也很好。查询都工作得很好，除了在几次之后才开始抛出这个错误。

Answer 1

"I am getting a leak detection every time I execute a query".

你当然是。在您的示例中，您从 DataSource 中得到 Connection，执行 PreparedStatement，关闭 PreparedStatement 然后不关闭 Connection，所以它是没有返回到池中并导致泄漏。

关闭你的人脉！只有您可以防止 fores...err，连接泄漏。

一定数量的查询后 HikariCP 连接超时错误

HikariCP connection timeout error after certain amount of queries

java

mysql

memory-leaks

jdbc

hikaricp