[
https://jira.nuxeo.com/browse/NXP-7956?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=106596#comment-106596
]
Olivier Grisel commented on NXP-7956:
-------------------------------------
By tracing the executions is seems that this is caused by the delayed full-time
indexing of documents created in the beforeMethodRun or beforeRun and then
deleted in the afterRun or afterMethodRun: the insertion of the extracted text
seems to occur after the document has been deleted by the cleanupSession method.
> Random crash in async event listeners on CoreFeature annotated tests with a
> repository init
> -------------------------------------------------------------------------------------------
>
> Key: NXP-7956
> URL: https://jira.nuxeo.com/browse/NXP-7956
> Project: Nuxeo Enterprise Platform
> Issue Type: Bug
> Components: Core SQL Storage
> Affects Versions: 5.4.2-SNAPSHOT
> Reporter: Olivier Grisel
> Assignee: Olivier Grisel
> Fix For: 5.5
>
>
> It seems that the H2 database does not handle concurrent writes on the
> fulltext table correctly and that can cause referential integrity related
> crashes when running the tests.
> I attach the typical stacktrace as first comment to this issue.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
_______________________________________________
ECM-tickets mailing list
[email protected]
http://lists.nuxeo.com/mailman/listinfo/ecm-tickets
