A very similar report came in a couple months ago, and the reason was that the session (/children) was incorrectly being used from multiple threads, causing responses operations to 'overlap' and leading to a ClassCastException when the mismatching responses for a different operation arrived.
On Thu, 13 Oct 2022 at 20:36, Justin Bertram <jbert...@apache.org> wrote: > > If you are able to reproduce with 2.26.0 please follow up with a Jira [1], > and attach the test you're using to reproduce it (or some other equivalent > way to reproduce the CCE). Thanks! > > > Justin > > [1] https://issues.apache.org/jira/browse/ARTEMIS > > On Thu, Oct 13, 2022 at 2:31 PM Thorsten Meinl <thorsten.me...@knime.com> > wrote: > > > Am Donnerstag, dem 13.10.2022 um 11:38 -0500 schrieb Justin Bertram: > > > It looks like perhaps the responses are being read by the client out > > > of order or something like that. Are you using both the core JMS > > > client and broker from 2.24.0? > > Yes, both client and broker are 2.24.0. > > > > > How often do you see this? Have you tried reproducing with the latest > > > release (i.e. 2.26.0)? > > Very infrequently actually. We have automated tests and they fail once > > every few days with this message. So in less than 1 out of 1000 I would > > say (very rough guess). > > > > I can try with 2.26.0. > > > > > > > On Thu, Oct 13, 2022 at 9:51 AM Thorsten Meinl > > > <thorsten.me...@knime.com> > > > wrote: > > > > > > > Hi all, > > > > > > > > Very occasionally we get the following exception when creating a > > > > new > > > > topic in Artemis (2.24.0): > > > > > > > > Caused by: java.lang.ClassCastException: class > > > > org.apache.activemq.artemis.core.protocol.core.impl.wireformat.Null > > > > Resp > > > > onseMessage_V2 cannot be cast to class > > > > org.apache.activemq.artemis.core.protocol.core.impl.wireformat.Sess > > > > ionB > > > > indingQueryResponseMessage_V4 > > > > (org.apache.activemq.artemis.core.protocol.core.impl.wireformat.Nul > > > > lRes > > > > ponseMessage_V2 and > > > > org.apache.activemq.artemis.core.protocol.core.impl.wireformat.Sess > > > > ionB > > > > indingQueryResponseMessage_V4 are in unnamed module of loader > > > > io.quarkus.bootstrap.classloading.QuarkusClassLoader @3bf0cb2d) > > > > at > > > > org.apache.activemq.artemis.core.protocol.core.impl.ActiveMQSession > > > > Cont > > > > ext.addressQuery(ActiveMQSessionContext.java:413) > > > > at > > > > org.apache.activemq.artemis.core.client.impl.ClientSessionImpl.addr > > > > essQ > > > > uery(ClientSessionImpl.java:794) > > > > at > > > > org.apache.activemq.artemis.jms.client.ActiveMQSession.lookupTopic( > > > > Acti > > > > veMQSession.java:1325) > > > > at > > > > org.apache.activemq.artemis.jms.client.ActiveMQSession.internalCrea > > > > teTo > > > > pic(ActiveMQSession.java:554) > > > > at > > > > org.apache.activemq.artemis.jms.client.ActiveMQSession.createTopic( > > > > Acti > > > > veMQSession.java:542) > > > > at > > > > org.apache.activemq.artemis.jms.client.ActiveMQJMSContext.createTop > > > > ic(A > > > > ctiveMQJMSContext.java:418) > > > > > > > > Any hints what could be the problem? > > > > > > > > Thanks, > > > > > > > > Thorsten > > > > > > > > -- > > > > Dr.-Ing. Thorsten Meinl > > > > KNIME AG > > > > Talacker 50 > > > > 8001 Zurich, Switzerland > > > > > > > > > > > > > > > > -- > > Dr.-Ing. Thorsten Meinl > > KNIME AG > > Talacker 50 > > 8001 Zurich, Switzerland > > > > > >
