C3P0 connection pool stall with Heroku Postgres

Question

This relates to a previously unsolved post: here C3P0 seems to enter into a deadlock immediately after initializing connection pool on Heroku. The problem does not occur on my local postgres.

Here's the connection configuration from hibernate.hbm.xml

 <property name="hibernate.dialect">org.hibernate.dialect.PostgreSQLDialect</property>



    <!-- Enable Hibernate's automatic session context management -->
    <property name="current_session_context_class">thread</property>


    <property name="hibernate.connection.url">jdbc:postgresql://myurl:5432/mydb?user=myusername&amp;password=myoassword&amp;sslfactory=org.postgresql.ssl.NonValidatingFactory&amp;ssl=true</property>
    <property name="hibernate.connection.driver_class">org.postgresql.Driver</property>
    <property name="hibernate.connection.username">myusername</property>
    <property name="hibernate.connection.password">mypassword</property>
    <property name="hibernate.archive.autodetection">class</property>
    <property name="hibernate.show_sql">true</property>
    <property name="hibernate.format_sql">true</property>
    <property name="hbm2ddl.auto">create</property>

    <!-- c3p0 connection pool settings -->
    <property name="hibernate.connection.provider_class">org.hibernate.service.jdbc.connections.internal.C3P0ConnectionProvider</property>
    <property name="hibernate.c3p0.min_size">1</property>
    <property name="hibernate.c3p0.max_size">2</property>
    <property name="hibernate.c3p0.timeout">1800</property>

Here's the log. Note the hang for about one minute with 10 second cycles of "Running DeadlockDetector." Also note the invocation exception that follows:

[DEBUG] "com.mchange.v2.c3p0.impl.C3P0PooledConnectionPoolManager" 2015-07-05 07:12:57,851: Created new pool for auth, username (masked): 'sp******'.

[DEBUG] "com.mchange.v2.resourcepool.BasicResourcePool" 2015-07-05 07:12:57,851: acquire test -- pool size: 0; target_pool_size: 1; desired target? 1

...deleted some lines for sake of parsimony

[DEBUG] "com.mchange.v2.resourcepool.BasicResourcePool" 2015-07-05 07:12:59,132: Acquisition series terminated successfully. Decremented pending_acquires 1, attempts_remaining: 30

[DEBUG] "com.mchange.v2.resourcepool.BasicResourcePool" 2015-07-05 07:12:59,132: trace com.mchange.v2.resourcepool.BasicResourcePool@6293abcc [managed: 2, unused: 1, excluded: 0] (e.g. com.mchange.v2.c3p0.impl.NewPooledConnection@576ddf03)

[DEBUG] "com.mchange.v2.resourcepool.BasicResourcePool" 2015-07-05 07:12:59,132: decremented pending_acquires: 0

[DEBUG] "com.mchange.v2.resourcepool.BasicResourcePool" 2015-07-05 07:12:59,132: Acquisition series terminated successfully. Decremented pending_acquires [0], attempts_remaining: 30

[DEBUG] "com.mchange.v2.resourcepool.BasicResourcePool" 2015-07-05 07:12:59,132: trace com.mchange.v2.resourcepool.BasicResourcePool@6293abcc [managed: 2, unused: 1, excluded: 0] (e.g. com.mchange.v2.c3p0.impl.NewPooledConnection@576ddf03)

[DEBUG] "com.mchange.v2.async.ThreadPoolAsynchronousRunner" 2015-07-05 07:13:07,841: com.mchange.v2.async.ThreadPoolAsynchronousRunner$DeadlockDetector@45912fb9 -- Running DeadlockDetector[Exiting. No pending tasks.]

...repeats every ten seconds

[DEBUG] "com.mchange.v2.async.ThreadPoolAsynchronousRunner" 2015-07-05 07:13:47,859: com.mchange.v2.async.ThreadPoolAsynchronousRunner$DeadlockDetector@45912fb9 -- Running DeadlockDetector[Exiting. No pending tasks.]

[INFO] "org.hibernate.dialect.Dialect" 2015-07-05 07:13:55,467: HHH000400: Using dialect: org.hibernate.dialect.PostgreSQLDialect

[INFO] "org.hibernate.engine.jdbc.internal.LobCreatorBuilder" 2015-07-05 07:13:55,472: HHH000424: Disabling contextual LOB creation as createClob() method threw error : java.lang.reflect.InvocationTargetException

[DEBUG] "com.mchange.v2.async.ThreadPoolAsynchronousRunner" 2015-07-05 07:13:55,576: com.mchange.v2.async.ThreadPoolAsynchronousRunner@6a01e23: Adding task to queue -- com.mchange.v2.resourcepool.BasicResourcePool$1RefurbishCheckinResourceTask@55b699ef

switched hibernate logging to debug. Got one line immediately before the deadlockdetector: [DEBUG] "org.hibernate.engine.jdbc.cursor.internal.StandardRefCursorSupport" 2015-07-05 07:39:21,781: Unexpected error trying to gauge level of JDBC REF_CURSOR support : null — Jake, Jul 05 '15 at 14:41

score 1 · Answer 1 · edited May 23 '17 at 12:21

1

The answer is, in fact, already posted and very well explained here:

I added the following to my hibernate.hbm.xml

<property name="temp.use_jdbc_metadata_defaults">false</property>

edited May 23 '17 at 12:21

Community

1
1

answered Jul 05 '15 at 14:54

Jake

4,322
6
39
83

C3P0 connection pool stall with Heroku Postgres

1 Answers1