1.1 错误信息：
Caused by: com.mysql.jdbc.exceptions.jdbc4.CommunicationsException: The last packet successfully received from the server was 20,820,001 milliseconds ago. The last packet sent successfully to the server was 20,820,002 milliseconds ago. is longer than the server configured value of 'wait_timeout'. You should consider either expiring and/or testing connection validity before use in your application, increasing the server configured values for client timeouts, or using the Connector/J connection property 'autoReconnect=true' to avoid this problem.
at sun.reflect.GeneratedConstructorAccessor29.newInstance(Unknown Source) ~[na:na]
at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45) ~[na:1.7.0_51]
at java.lang.reflect.Constructor.newInstance(Constructor.java:526) ~[na:1.7.0_51]
at com.mysql.jdbc.Util.handleNewInstance(Util.java:411) ~[mysql-connector-java-5.1.29.jar:na]
at com.mysql.jdbc.SQLError.createCommunicationsException(SQLError.java:1129) ~[mysql-connector-java-5.1.29.jar:na]
at com.mysql.jdbc.MysqlIO.send(MysqlIO.java:3988) ~[mysql-connector-java-5.1.29.jar:na]
at com.mysql.jdbc.MysqlIO.sendCommand(MysqlIO.java:2598) ~[mysql-connector-java-5.1.29.jar:na]
at com.mysql.jdbc.MysqlIO.sqlQueryDirect(MysqlIO.java:2778) ~[mysql-connector-java-5.1.29.jar:na]
at com.mysql.jdbc.ConnectionImpl.execSQL(ConnectionImpl.java:2828) ~[mysql-connector-java-5.1.29.jar:na]
at com.mysql.jdbc.ConnectionImpl.setAutoCommit(ConnectionImpl.java:5372) ~[mysql-connector-java-5.1.29.jar:na]
at com.mchange.v2.c3p0.impl.NewProxyConnection.setAutoCommit(NewProxyConnection.java:881) ~[c3p0-0.9.1.1.jar:0.9.1.1]
at org.quartz.impl.jdbcjobstore.AttributeRestoringConnectionInvocationHandler.setAutoCommit(AttributeRestoringConnectionInvocationHandler.java:98) ~[quartz-2.2.1.jar:na]
1.2 解决方法
- 如果使用的是JDBC，在JDBC URL上添加?autoReconnect=true，如：
jdbc:mysql://10.10.10.10:3306/mydb?autoReconnect=true
- 如果是在Spring中使用DBCP连接池，在定义datasource增加属性validationQuery和testOnBorrow，如：
<bean id="vrsRankDataSource" class="org.apache.commons.dbcp.BasicDataSource" destroy-method="close">
<property name="driverClassName" value="${jdbc.driverClassName}" />
<property name="url" value="${countNew.jdbc.url}" />
<property name="username" value="${countNew.jdbc.user}" />
<property name="password" value="${countNew.jdbc.pwd}" />
<property name="validationQuery" value="SELECT 1" />
<property name="testOnBorrow" value="true"/>
</bean>- 如果是在Spring中使用c3p0连接池，则在定义datasource的时候，添加属性testConnectionOnCheckin和testConnectionOnCheckout，如：
<bean name="cacheCloudDB" class="com.mchange.v2.c3p0.ComboPooledDataSource">
<property name="driverClass" value="${jdbc.driver}"/>
<property name="jdbcUrl" value="${cache.url}"/>
<property name="user" value="${cache.user}"/>
<property name="password" value="${cache.password}"/>
<property name="initialPoolSize" value="10"/>
<property name="maxPoolSize" value="${cache.maxPoolSize}"/>
<property name="testConnectionOnCheckin" value="false"/>
<property name="testConnectionOnCheckout" value="true"/>
<property name="preferredTestQuery" value="SELECT 1"/>
</bean>参考
附录分析
When a c3p0-proxied Connection throws an SQLException, c3p0 examines
the Exception and the Connection to make a judgement about whether
the problem implies that the Connection should no longer be included
in the pool. c3p0 tests the Connection, and if the test fails, the
Connection will be excluded from the pool.
What c3p0 is telling you here is that a Connection that previously
signalled an error and then failed a Connection test is still in use,
and has signalled another error. From c3p0's perspective, this is a
non-issue, it just means c3p0 doesn't have to do any kind of checks
or notifications, the Connection is already gone as far as the pool
is concerned. But c3p0 wonders why you'd still be using such a
Connection, and warns you about it.
Usually, if a client continues to use a Connection that c3p0 has
correctly identified as broken, all further uses will provoke such an
exception, and the fix is to close the Connection and start over when
an application's Connection turns out to be dead. But, by the error
you're getting, it looks like your Connection is still live and okay
-- it's clearly communicating with the database. So, the issue is,
why did c3p0 deem the Connection dead if it is not?
If you turn on DEBUG level logging (relevant loggers would be
com.mchange.v2.c3p0.impl.NewPooledConnection,
com.mchange.v2.c3p0.impl.C3P0PooledConnectionPool, and
com.mchange.v2.c3p0.impl.DefaultConnectionTester, unless you've
defined your own ConnectionTester), you can trace the testing and
invalidation of Connections, and try to understand why Connections
that seem okay are testing as broken. That will give you better
information about what's going on.
That said, the only cost of this behavior is disconcerting warning
messages and somewhat faster churn of Connections through the pool.
c3p0 is erring on the side of caution -- it has reason to believe a
Connection is bad, so it's been excluded from the pool. It'd be nice
to know why apparently good Connections are failing Connection tests,
but if it is an infrequent occurrence, it's very little to worry
about. (If it's happening a lot, you should track it down.)
原文地址：http://nkcoder.github.io/blog/20140712/mysql-reconnect-packet-lost/
原创文章如转载，请注明出处“
伊人博客”