We're using Keycloak 1.9.0.Final as of now. I had a quick check on our logs and it seems that we have performed update calls to the same user almost at the same time when this error has popped up. Could be a programmatical error on one of our clients.

In any case, the occurrences of this error are very rare. Just wanted to note it down here for your exposure. Thanks for the response.


On Fri, Mar 18, 2016 at 2:49 PM, Marek Posolda <mposolda@redhat.com> wrote:

It seems that you're trying to update same user by 2 concurrent transactions, hence there is OptimisticLockException thrown by Hibernate.

On which version are you? We had some cache fixes for user cache in latest master, but that's related to infinispan cache. You can try with latest master, but not sure if it helps with the scenario you're testing as the error is at DB level. Not sure if Keycloak should somehow improve the support of the scanario when user is updated by 2 transactions at the same time. Doesn't look like very common thing to me.


On 17/03/16 11:03, Lohitha Chiranjeewa wrote:

We were executing the 'Update User' API call when the following exception got thrown:
[2016-03-17 07:50:36.0590], DEBUG, org.jboss.resteasy.core.SynchronousDispatcher default task-13 - RESTEASY002315: PathInfo: /admin/realms/xxxx/users/e503cb04-3080-4e90-a4b4-80adcd46b81c
[2016-03-17 07:50:36.0590], DEBUG, org.keycloak.services.managers.AuthenticationManager default task-13 - token active - active: true, issued-at: 1,458,201,036, not-before: 0
[2016-03-17 07:50:36.0590], DEBUG, org.keycloak.services.resources.admin.AdminRoot default task-13 - authenticated admin access for: superuser
[2016-03-17 07:50:36.0591], DEBUG, org.keycloak.services.resources.Cors default task-13 - No origin returning
[2016-03-17 07:50:36.0636], DEBUG, org.infinispan.interceptors.InvalidationInterceptor default task-13 - Cache [localhost] replicating InvalidateCommand{keys=[e503cb04-3080-4e90-a4b4-80adcd46b81c]}
[2016-03-17 07:50:36.0637], ERROR, org.keycloak.services.resources.ModelExceptionMapper default task-13 - javax.persistence.OptimisticLockException: Batch update returned unexpected row count from update [0]; actual row count: 0; expected: 1: org.keycloak.models.ModelException: javax.persistence.OptimisticLockException: Batch update returned unexpected row count from update [0]; actual row count: 0; expected: 1
    at org.keycloak.connections.jpa.PersistenceExceptionConverter.convert(PersistenceExceptionConverter.java:61)
    at org.keycloak.connections.jpa.JpaKeycloakTransaction.commit(JpaKeycloakTransaction.java:47)
    at org.keycloak.services.DefaultKeycloakTransactionManager.commit(DefaultKeycloakTransactionManager.java:95)
    at org.keycloak.services.resources.admin.UsersResource.updateUser(UsersResource.java:170)
    at sun.reflect.GeneratedMethodAccessor435.invoke(Unknown Source)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:497)
    at org.jboss.resteasy.core.MethodInjectorImpl.invoke(MethodInjectorImpl.java:139)
    at org.jboss.resteasy.core.ResourceMethodInvoker.invokeOnTarget(ResourceMethodInvoker.java:295)
    at org.jboss.resteasy.core.ResourceMethodInvoker.invoke(ResourceMethodInvoker.java:249)
    at org.jboss.resteasy.core.ResourceLocatorInvoker.invokeOnTargetObject(ResourceLocatorInvoker.java:138)
    at org.jboss.resteasy.core.ResourceLocatorInvoker.invoke(ResourceLocatorInvoker.java:107)
We're using a clustered Infinispan system as our cache. This error doesn't get thrown often, nor can we point out a specific scenario. However, we have now seen this a few times. What could be the reason for this?


keycloak-user mailing list