[
https://issues.apache.org/jira/browse/OPENJPA-1767?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12899966#action_12899966
]
Rick Curtis commented on OPENJPA-1767:
--------------------------------------
After further research I believe the problem is more widespread that just
OneToMany relationships. The problem is when committing data to the datacache
we need to check if the owning side of a relationship has a dirty relationship
field. If that field is dirty we need to ensure that the referenced side of the
relationship is kicked out of the cache.
> OneToMany relationships owned by the many side aren't properly evicted from
> the DataCache
> -----------------------------------------------------------------------------------------
>
> Key: OPENJPA-1767
> URL: https://issues.apache.org/jira/browse/OPENJPA-1767
> Project: OpenJPA
> Issue Type: Bug
> Components: datacache
> Affects Versions: 2.0.0, 2.1.0
> Reporter: Rick Curtis
> Priority: Critical
>
> When working on OPENJPA-1469 I found that the test
> org.apache.openjpa.persistence.spring.TestLibService fails because we have
> stale data in the cache. To recreate this failure, set
> "openjpa.DataCache","true" in the setup() method.
> The net of the problem is that if the non-owning side of the relationship
> exists in the datacache and the relationship is updated on the owning side,
> the non-owning side doesn't get updated.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
