Hibernate Inconsistent and Conflicting Results with Merge

Question

VERSION Red Hat Build of Quarkus

  <properties>
    <compiler-plugin.version>3.8.1</compiler-plugin.version>
    <maven.compiler.parameters>true</maven.compiler.parameters>
    <maven.compiler.source>11</maven.compiler.source>
    <maven.compiler.target>11</maven.compiler.target>
    <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
    <project.reporting.outputEncoding>UTF-8</project.reporting.outputEncoding>
    <quarkus-plugin.version>1.7.5.Final</quarkus-plugin.version>
    <quarkus.platform.artifact-id>quarkus-universe-bom</quarkus.platform.artifact-id>
    <quarkus.platform.group-id>io.quarkus</quarkus.platform.group-id>
    <quarkus.platform.version>1.7.5.Final</quarkus.platform.version>
    <surefire-plugin.version>3.0.0-M5</surefire-plugin.version>
  </properties>

I am totally confused about this inconsistency and despite logging set to the maximum, all I see is that my code is not behaving the same in this project as compared to other projects.

Everything is configured the same for versions and settings.

I have three quarkus jax-rs services with hibernate-orm pushing to PostgreSQL.

2 of the services all behave as expected with approximately 25 identically configured PUT endpoints with a merge backing manager (code below). In the two working services, the statement in the logs shows where an UPDATE statement is required. However, the third (new) service simply ignores the change. It responds with 200 and even returns the updated data, but the database is never updated.

simple entity

@Entity
@Table(name = "applicants")
public class Applicant implements Serializable {

    private static final long serialVersionUID = 6796452320678418766L;

    @Column(name = "applicant_id")
    @Id
    @GeneratedValue(strategy = GenerationType.SEQUENCE, generator = "applicant_seq")
    public long applicant_id;

    @Column(name = "email", nullable = false, unique = true)
    public String email;
}

@PUT Resource

@PUT
    @Path("{id}")
    public Response update(@PathParam("id") Long id, @Valid Applicant applicant) {
        try {
            Applicant applicantToUpdate = this.manager.findById(id);
            if (applicantToUpdate != null) {
                applicantToUpdate = this.manager.save(applicant);
                if (applicantToUpdate != null) {
                    return Response.ok(applicantToUpdate).build();
                } else {
                    ...
                }
            }
            return Response.status(Response.Status.NOT_FOUND).build();
        } catch (Exception ex) {
            ...
        }
    }

Backing Manager for persistance

@Transactional
public Applicant save(Applicant applicant) {
    return this.em.merge(applicant);
}

Both the ApplicantResource and the ApplicantManager classes are @RequestScoped.

The logs for the failing service

2020-11-16 13:51:26,122 DEBUG [org.hib.res.jdb.int.LogicalConnectionManagedImpl] (executor-thread-198) `hibernate.connection.provider_disables_autocommit` was enabled.  This setting should only be enabled when you are certain that the Connections given to Hibernate by the ConnectionProvider have auto-commit disabled.  Enabling this setting when the Connections do not have auto-commit disabled will lead to Hibernate executing SQL operations outside of any JDBC/SQL transaction.
2020-11-16 13:51:26,123 DEBUG [org.hib.SQL] (executor-thread-198) 
    select
        applicant0_.applicant_id as applican1_0_0_,
        applicant0_.email as email2_0_0_
    from
        applicants applicant0_
    where
        applicant0_.applicant_id=?
Hibernate: 
    select
        applicant0_.applicant_id as applican1_0_0_,
        applicant0_.email as email2_0_0_
    from
        applicants applicant0_
    where
        applicant0_.applicant_id=?
2020-11-16 13:51:26,156 DEBUG [org.hib.loa.pla.exe.pro.int.EntityReferenceInitializerImpl] (executor-thread-198) On call to EntityIdentifierReaderImpl#resolve, EntityKey was already known; should only happen on root returns with an optional identifier specified
2020-11-16 13:51:26,156 DEBUG [org.hib.eng.int.TwoPhaseLoad] (executor-thread-198) Resolving attributes for [org.protechskillsinstitute.applicant.entity.Applicant#1]
2020-11-16 13:51:26,157 DEBUG [org.hib.eng.int.TwoPhaseLoad] (executor-thread-198) Processing attribute `email` : value = original@email.com
2020-11-16 13:51:26,157 DEBUG [org.hib.eng.int.TwoPhaseLoad] (executor-thread-198) Attribute (`email`)  - enhanced for lazy-loading? - false
2020-11-16 13:51:26,157 DEBUG [org.hib.eng.int.TwoPhaseLoad] (executor-thread-198) Done materializing entity [org.protechskillsinstitute.applicant.entity.Applicant#1]
2020-11-16 13:51:26,159 DEBUG [org.hib.res.jdb.int.LogicalConnectionManagedImpl] (executor-thread-198) Initiating JDBC connection release from afterStatement
2020-11-16 13:51:26,159 DEBUG [org.hib.loa.ent.pla.AbstractLoadPlanBasedEntityLoader] (executor-thread-198) Done entity load : org.protechskillsinstitute.applicant.entity.Applicant#1
2020-11-16 13:51:26,159 DEBUG [org.hib.res.jdb.int.LogicalConnectionManagedImpl] (executor-thread-198) Initiating JDBC connection release from afterTransaction
2020-11-16 13:51:26,160 DEBUG [org.hib.res.jdb.int.LogicalConnectionManagedImpl] (executor-thread-198) `hibernate.connection.provider_disables_autocommit` was enabled.  This setting should only be enabled when you are certain that the Connections given to Hibernate by the ConnectionProvider have auto-commit disabled.  Enabling this setting when the Connections do not have auto-commit disabled will lead to Hibernate executing SQL operations outside of any JDBC/SQL transaction.
2020-11-16 13:51:26,160 DEBUG [org.hib.res.tra.bac.jta.int.JtaTransactionCoordinatorImpl] (executor-thread-198) Hibernate RegisteredSynchronization successfully registered with JTA platform
2020-11-16 13:51:26,161 DEBUG [org.hib.res.tra.bac.jta.int.JtaTransactionCoordinatorImpl] (executor-thread-198) JTA transaction was already joined (RegisteredSynchronization already registered)
2020-11-16 13:51:26,161 DEBUG [org.hib.eve.int.EntityCopyObserverFactoryInitiator] (executor-thread-198) Configured EntityCopyObserver strategy: disallow
2020-11-16 13:51:26,161 DEBUG [org.hib.loa.Loader] (executor-thread-198) Loading entity: [org.protechskillsinstitute.applicant.entity.Applicant#1]
2020-11-16 13:51:26,212 DEBUG [org.hib.SQL] (executor-thread-198) 
    select
        applicant0_.applicant_id as applican1_0_0_,
        applicant0_.email as email2_0_0_
    from
        applicants applicant0_
    where
        applicant0_.applicant_id=?
Hibernate:
    select
        applicant0_.applicant_id as applican1_0_0_,
        applicant0_.email as email2_0_0_
    from
        applicants applicant0_
    where
        applicant0_.applicant_id=?
2020-11-16 13:51:26,216 DEBUG [org.hib.loa.Loader] (executor-thread-198) Result set row: 0
2020-11-16 13:51:26,216 DEBUG [org.hib.loa.Loader] (executor-thread-198) Result row: EntityKey[org.protechskillsinstitute.applicant.entity.Applicant#1]
2020-11-16 13:51:26,217 DEBUG [org.hib.eng.int.TwoPhaseLoad] (executor-thread-198) Resolving attributes for [org.protechskillsinstitute.applicant.entity.Applicant#1]
2020-11-16 13:51:26,217 DEBUG [org.hib.eng.int.TwoPhaseLoad] (executor-thread-198) Processing attribute `email` : value = original@email.com
2020-11-16 13:51:26,217 DEBUG [org.hib.eng.int.TwoPhaseLoad] (executor-thread-198) Attribute (`email`)  - enhanced for lazy-loading? - false
2020-11-16 13:51:26,217 DEBUG [org.hib.eng.int.TwoPhaseLoad] (executor-thread-198) Done materializing entity [org.protechskillsinstitute.applicant.entity.Applicant#1]
2020-11-16 13:51:26,218 DEBUG [org.hib.res.jdb.int.LogicalConnectionManagedImpl] (executor-thread-198) Initiating JDBC connection release from afterStatement
2020-11-16 13:51:26,218 DEBUG [org.hib.loa.Loader] (executor-thread-198) Done entity load
2020-11-16 13:51:26,219 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Processing flush-time cascades
2020-11-16 13:51:26,219 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Dirty checking collections
2020-11-16 13:51:26,219 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Flushed: 0 insertions, 0 updates, 0 deletions to 1 objects
2020-11-16 13:51:26,220 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Flushed: 0 (re)creations, 0 updates, 0 removals to 0 collections
2020-11-16 13:51:26,220 DEBUG [org.hib.int.uti.EntityPrinter] (executor-thread-198) Listing entities:
2020-11-16 13:51:26,220 DEBUG [org.hib.int.uti.EntityPrinter] (executor-thread-198) org.protechskillsinstitute.applicant.entity.Applicant{applicant_id=1, email=stephen.w.boyd@gmail.com}
2020-11-16 13:51:26,220 DEBUG [org.hib.res.jdb.int.LogicalConnectionManagedImpl] (executor-thread-198) Initiating JDBC connection release from afterStatement
2020-11-16 13:51:26,221 DEBUG [org.hib.eng.tra.int.TransactionImpl] (executor-thread-198) On TransactionImpl creation, JpaCompliance#isJpaTransactionComplianceEnabled == false
2020-11-16 13:51:26,221 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Processing flush-time cascades
2020-11-16 13:51:26,221 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Dirty checking collections
2020-11-16 13:51:26,221 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Flushed: 0 insertions, 0 updates, 0 deletions to 1 objects
2020-11-16 13:51:26,222 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Flushed: 0 (re)creations, 0 updates, 0 removals to 0 collections
2020-11-16 13:51:26,222 DEBUG [org.hib.int.uti.EntityPrinter] (executor-thread-198) Listing entities:
2020-11-16 13:51:26,222 DEBUG [org.hib.int.uti.EntityPrinter] (executor-thread-198) org.protechskillsinstitute.applicant.entity.Applicant{applicant_id=1, email=edited@email.com}
2020-11-16 13:51:26,222 DEBUG [org.hib.res.jdb.int.LogicalConnectionManagedImpl] (executor-thread-198) Initiating JDBC connection release from afterStatement
2020-11-16 13:51:26,223 INFO  [org.hib.eng.int.StatisticalLoggingSessionEventListener] (executor-thread-198) Session Metrics {
    112700 nanoseconds spent acquiring 1 JDBC connections;
    9700 nanoseconds spent releasing 1 JDBC connections;
    147400 nanoseconds spent preparing 1 JDBC statements;
    340500 nanoseconds spent executing 1 JDBC statements;
    0 nanoseconds spent executing 0 JDBC batches;
    0 nanoseconds spent performing 0 L2C puts;
    0 nanoseconds spent performing 0 L2C hits;
    0 nanoseconds spent performing 0 L2C misses;
    3546700 nanoseconds spent executing 2 flushes (flushing a total of 2 entities and 0 collections);
    0 nanoseconds spent executing 0 partial-flushes (flushing a total of 0 entities and 0 collections)
}
2020-11-16 13:51:26,227 DEBUG [org.hib.res.jdb.int.LogicalConnectionManagedImpl] (executor-thread-198) Initiating JDBC connection release from afterTransaction
2020-11-16 13:51:26,230 INFO  [org.hib.eng.int.StatisticalLoggingSessionEventListener] (executor-thread-198) Session Metrics {
    34900 nanoseconds spent acquiring 1 JDBC connections;
    18700 nanoseconds spent releasing 1 JDBC connections;
    77500 nanoseconds spent preparing 1 JDBC statements;
    430200 nanoseconds spent executing 1 JDBC statements;
    0 nanoseconds spent executing 0 JDBC batches;
    0 nanoseconds spent performing 0 L2C puts;
    0 nanoseconds spent performing 0 L2C hits;
    0 nanoseconds spent performing 0 L2C misses;
    0 nanoseconds spent executing 0 flushes (flushing a total of 0 entities and 0 collections);
    0 nanoseconds spent executing 0 partial-flushes (flushing a total of 0 entities and 0 collections)
}

Everything in the code and logs are identical for the working and non-working services with one exception. The log shows the Update is required for the working services and does not for the failing service.

Thoughts? As always, thank you for your assistance!

Drilled in logs

2020-11-16 13:51:26,156 DEBUG [org.hib.loa.pla.exe.pro.int.EntityReferenceInitializerImpl] (executor-thread-198) On call to EntityIdentifierReaderImpl#resolve, EntityKey was already known; should only happen on root returns with an optional identifier specified
2020-11-16 13:51:26,156 DEBUG [org.hib.eng.int.TwoPhaseLoad] (executor-thread-198) Resolving attributes for [org.protechskillsinstitute.applicant.entity.Applicant#1]
2020-11-16 13:51:26,157 DEBUG [org.hib.eng.int.TwoPhaseLoad] (executor-thread-198) Processing attribute `email` : value = original@email.com
2020-11-16 13:51:26,157 DEBUG [org.hib.eng.int.TwoPhaseLoad] (executor-thread-198) Attribute (`email`)  - enhanced for lazy-loading? - false

and the check

11-16 13:51:26,220 DEBUG [org.hib.int.uti.EntityPrinter] (executor-thread-198) Listing entities:
2020-11-16 13:51:26,220 DEBUG [org.hib.int.uti.EntityPrinter] (executor-thread-198) org.protechskillsinstitute.applicant.entity.Applicant{applicant_id=1, email=edited@email.com}
2020-11-16 13:51:26,220 DEBUG [org.hib.res.jdb.int.LogicalConnectionManagedImpl] (executor-thread-198) Initiating JDBC connection release from afterStatement
2020-11-16 13:51:26,221 DEBUG [org.hib.eng.tra.int.TransactionImpl] (executor-thread-198) On TransactionImpl creation, JpaCompliance#isJpaTransactionComplianceEnabled == false
2020-11-16 13:51:26,221 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Processing flush-time cascades
2020-11-16 13:51:26,221 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Dirty checking collections
2020-11-16 13:51:26,221 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Flushed: 0 insertions, 0 updates, 0 deletions to 1 objects
2020-11-16 13:51:26,222 DEBUG [org.hib.eve.int.AbstractFlushingEventListener] (executor-thread-198) Flushed: 0 (re)creations, 0 updates, 0 removals to 0 collections
2020-11-16 13:51:26,222 DEBUG [org.hib.int.uti.EntityPrinter] (executor-thread-198) Listing entities:
2020-11-16 13:51:26,222 DEBUG [org.hib.int.uti.EntityPrinter] (executor-thread-198) org.protechskillsinstitute.applicant.entity.Applicant{applicant_id=1, email=edited@email.com}

Answer 1

Are you sure the email actually changed? Hibernate will obviously not execute an update query if the data didn't change.

Answer 2

Based on the Resolved Dependencies it looks like Red Hat is using:

1. hibernate-core: 5.4.21.Final-redhat-00001
2. hibarenate-commons-annotations: 5.1.0.Final-redhat-00005
3. hibernate-graalvm: 5.4.21.Final-redhat-00001

Per the comments below, I have opened a ticket with Red Hat.

Case 02805123

Answer 3

Have you tried flush and commit after merge? did you check if the flush mode is manual or auto?