Hi, I noticed the same behaviour in my aerogear test server, but I thought that was something related with openshift. If service is not used for a while seems like it become suspended until you access again.

I tried to reproduce restarting the server but I couldn't... Just found something in the logs this morning after 2 days without using it. (At 2:51 I sent the notification)

Log was
-----------------------------
2015/06/13 00:00:00,109 INFO  [GCMPushNotificationSender] (EJB default - 2) Deleting ['0'] outdated PushMessageInformation objects
2015/06/13 12:05:30,766 INFO  [org.jboss.as.osgi] (MSC service thread 1-5) JBAS011942: Stopping OSGi Framework
2015/06/15 02:51:14,161 INFO  [org.jboss.as.configadmin] (ServerService Thread Pool -- 35) JBAS016200: Activating ConfigAdmin Subsystem
2015/06/15 02:51:14,264 INFO  [org.jboss.as.jacorb] (ServerService Thread Pool -- 41) JBAS016300: Activating JacORB Subsystem
2015/06/15 02:51:14,356 INFO  [org.jboss.as.security] (ServerService Thread Pool -- 58) JBAS013101: Activating Security Subsystem
2015/06/15 02:51:14,258 INFO  [org.jboss.as.clustering.infinispan] (ServerService Thread Pool -- 40) JBAS010280: Activating Infinispan subsystem.
...
...
-------------------------------------
   
It seem like on saturday 13 service was stopped. And started again on monday 16 at 2:51 (by my notification???) . I also got this error during the startup process

--------------------------
2015/06/15 02:51:46,443 INFO  [org.keycloak.adapters.as7.KeycloakAuthenticatorValve] (MSC service thread 1-5) **** using /WEB-INF/keycloak.json
2015/06/15 02:51:46,760 INFO  [JPAPushMessageInformationDao] (EJB default - 2) Deleting ['2'] outdated PushMessageInformation objects
2015/06/15 02:51:46,951 INFO  [org.keycloak.services.resources.KeycloakApplication] (MSC service thread 1-1) Not importing realm aerogear from resource /WEB-INF/ups-realm.json.  It already exists.
2015/06/15 02:51:46,954 INFO  [JPAPushMessageInformationDao] (EJB default - 1) Deleting ['2'] outdated PushMessageInformation objects
2015/06/15 02:51:46,966 WARN  [com.arjuna.ats.arjuna] (EJB default - 1) ARJUNA012125: TwoPhaseCoordinator.beforeCompletion - failed for SynchronizationImple< 0:ffff0a21a655:16c5d4f8:557e75e5:f, org.hibernate.engine.transaction.synchronization.internal.RegisteredSynchronization@d8bbe6 >: javax.persistence.OptimisticLockException: org.hibernate.StaleStateException: Batch update returned unexpected row count from update [0]; actual row count: 0; expected: 1
    at org.hibernate.ejb.AbstractEntityManagerImpl.wrapStaleStateException(AbstractEntityManagerImpl.java:1402) [hibernate-entitymanager-4.0.1.Final.jar:4.0.1.Final]
    at org.hibernate.ejb.AbstractEntityManagerImpl.convert(AbstractEntityManagerImpl.java:1308) [hibernate-entitymanager-4.0.1.Final.jar:4.0.1.Final]
    at org.hibernate.ejb.AbstractEntityManagerImpl.convert(AbstractEntityManagerImpl.java:1289) [hibernate-entitymanager-4.0.1.Final.jar:4.0.1.Final]
    at org.hibernate.ejb.AbstractEntityManagerImpl.convert(AbstractEntityManagerImpl.java:1295) [hibernate-entitymanager-4.0.1.Final.jar:4.0.1.Final]
    at org.hibernate.ejb.AbstractEntityManagerImpl$CallbackExceptionMapperImpl.mapManagedFlushFailure(AbstractEntityManagerImpl.java:1481) [hibernate-entitymanager-4.0.1.Final.jar:4.0.1.Final]
    at org.hibernate.engine.transaction.synchronization.internal.SynchronizationCallbackCoordinatorImpl.beforeCompletion(SynchronizationCallbackCoordinatorImpl.java:109) [hibernate-core-4.0.1.Final.jar:4.0.1.Final]
    at org.hibernate.engine.transaction.synchronization.internal.RegisteredSynchronization.beforeCompletion(RegisteredSynchronization.java:53) [hibernate-core-4.0.1.Final.jar:4.0.1.Final]
------------------------

Later another generated message at 2:57 was received an processed fine. I don't know if it's something related with hosting or with UPS. I'm running 1.0.2 too.

Sorry if this information is not relevant

Alex Ballesté

El 13/06/15 a les 08:01, Matthias Wessendorf ha escrit: