0

I'm having some trouble trying to catch an exception when there are concurrency violations using hibernate and Spring AOP. This is my scenario:

(My MyConcurrentStateControl has a @Version column)

Service layer

 @Override
public Integer saveWork(WorkDto dto) throws MyException {
    try {
        return workBusinessLogicService.saveWork(dto);
    } catch (PersistenceException ex) {
        // -- Concurrent insert exception
        if (ex.getCause() instanceof ConstraintViolationException) {
            String errorMsg = "The same item is being created by another user. Please refresh.";
            LOGGER.error(errorMsg, ex);
            throw new MyException(errorMsg);
        }
    } catch (StaleObjectStateException ole){
        // -- Concurrent update exception
        String errorMsg = "The same item is being updated by another user. Please refresh";
        LOGGER.error(errorMsg, ole);
        throw new MyException(errorMsg);
    } 
    return -1;
}

Business Logic layer

@Override
@Transactional(rollbackFor = {Exception.class, MyException.class, PersistenceException.class, StaleObjectStateException.class})
public Integer saveWork(WorkDto dto) throws MyException, PersistenceException, StaleObjectStateException {

    MyConcurrentStateControl concurrentState = concurrentStateManager.getState(dto.getId());

    if (concurrentState == null) {
        concurrentState = new MyConcurrentStateControl();
    }

    // -- Do updates in some other tables --

    // Save Concurrent State for concurrency check (Optimistic Locking)       
    Integer id = concurrentStateManager.save(concurrentState);

    // -- Also tried entityManager.flush();

}

This is the error I'm getting in the log, and it is thrown in this line:

 return workBusinessLogicService.saveWork(dto);

It is the error I'm expecting when multiple threads call the service, but I can't do anything with it.

[#|2015-01-30T20:53:49.793+0000|WARNING|glassfish3.1.2|javax.enterprise.system.core.transaction.com.sun.jts.jta|_ThreadID=32;_ThreadName=Thread-12;|JTS5054: Unexpected error occurred in after completion
org.hibernate.StaleObjectStateException: Row was updated or deleted by another transaction (or unsaved-value mapping was incorrect): [com.mymodel.entity.ConcurrentState#7]
    at org.hibernate.persister.entity.AbstractEntityPersister.check(AbstractEntityPersister.java:2471)
    at org.hibernate.persister.entity.AbstractEntityPersister.update(AbstractEntityPersister.java:3123)
    at org.hibernate.persister.entity.AbstractEntityPersister.updateOrInsert(AbstractEntityPersister.java:3021)
    at org.hibernate.persister.entity.AbstractEntityPersister.update(AbstractEntityPersister.java:3350)
    at org.hibernate.action.internal.EntityUpdateAction.execute(EntityUpdateAction.java:140)
......................
org.springframework.transaction.UnexpectedRollbackException: JTA transaction unexpectedly rolled back (maybe due to a timeout); nested exception is javax.transaction.RollbackException
    at org.springframework.transaction.jta.JtaTransactionManager.doCommit(JtaTransactionManager.java:1012)
    at org.springframework.transaction.support.AbstractPlatformTransactionManager.processCommit(AbstractPlatformTransactionManager.java:754)
    at org.springframework.transaction.support.AbstractPlatformTransactionManager.commit(AbstractPlatformTransactionManager.java:723)
    at org.springframework.transaction.interceptor.TransactionAspectSupport.commitTransactionAfterReturning(TransactionAspectSupport.java:393)
    at org.springframework.transaction.interceptor.TransactionInterceptor.invoke(TransactionInterceptor.java:120)
    at org.springframework.aop.framework.ReflectiveMethodInvocation.proceed(ReflectiveMethodInvocation.java:172)
    at net.bull.javamelody.MonitoringSpringInterceptor.invoke(MonitoringSpringInterceptor.java:74)
    at org.springframework.aop.framework.ReflectiveMethodInvocation.proceed(ReflectiveMethodInvocation.java:172)
    at org.springframework.aop.framework.Cglib2AopProxy$DynamicAdvisedInterceptor.intercept(Cglib2AopProxy.java:621)
    at com.mymodel.business.core.service.WorkServiceImpl$$EnhancerByCGLIB$$b1800d24.saveWork(<generated>)

I know the actual queries are triggered on transaction commit, and at that point the control has moved out of the method, therefore I'm not being able to catch the StaleObjectStateException. But how can I do that, or is there some alternative? All I want is:

  1. Roll back all transactions.
  2. Show a reasonable message to the user that there have been concurrent updates and he needs to refresh the UI.
user3510607
  • 13
  • 4
  • You seem to be using JPA over hibernate. So I think a OptimisticLockException is being thrown instead of hibernate's StaleObjectException. Can you add a catch block for it before the PersistenceException block and confirm? – Andy Dufresne Jan 31 '15 at 11:56
  • Thanks Andy, as shown, the stacktrace throws StaleObjectStateException, we are using the Hibernate implementation of JPA. As described in this other post, we are locked into that implementation: http://stackoverflow.com/questions/18176932/staleobjectstateexception-vs-optimisticlockexception – user3510607 Feb 01 '15 at 21:47

0 Answers0