Hide Forgot
Date of First Response: 2007-12-18 11:08:06 project_key: SOA SOA-P platform JIRA for JBPAPP JIRA: http://jira.jboss.com/jira/browse/JBPAPP-479
Link: Added: This issue depends JBPAPP-479
Tom to clarify with Steve if jBPM is using Hibernate correctly. http://jira.jboss.com/jira/browse/JBPAPP-479#action_12392571
Tom, what's the status?
i analysed the status of the support case yesterday. now i need to see whether/how/why we generate this not-very-meaningful sequence of calls to the hibernate session.
Tom is going to look at this on the 28th of January.
Len, Do you recall the original scenario that revealed the problem in the context of jBPM. The referenced jira issue and the indirectly referenced issue https://enterprise.redhat.com/issue-tracker/?module=issues&action=view&tid=157838 only contain the minimal example that was rebuild to reproduce the problem. But I can't seem to find the description on how to reproduce the problem with jBPM.
len is going to try and reproduce the original jbpm problem. from that scenario, i should be able to trace it down. ...if the problem is still there.
<alexk> ldimaggi_brb: Fady say it was fixed. Can you look at https://na1.salesforce.com/50030000004BGPu Does anyone have access?
Link: Added: This issue depends JBQA-859
As per my test suite runs there are no failings with oracle on CR2 standalone. Though embedded contains a wrong version of the jar so it is not passing there. Anyway closing this issue as jar version is covered in SOA-359
What bugs me is that it was a hibernate issue, but we test with jBPM. So do we need to ensure hibernate was fixed or it is enough to know the issue doesn't affect jBPM? Len, please reopen if needed!
Aleksandar, Steve explained that this was intended behaviour of Hibernate and not really considered as a bug. The thing that is still strange is that this strange sequence of calls to hibernate would only be happening (or only leads to trouble) on oracle. That is weird, but possible. To resolve this, I would need the exact jBPM scenario to reproduce it. And then I can verify what calls are made to Hibernate and if that sequence matches the sequence described by Steve.