[
https://issues.apache.org/jira/browse/OAK-8353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16849538#comment-16849538
]
Marcel Reutegger edited comment on OAK-8353 at 5/28/19 12:39 PM:
-----------------------------------------------------------------
Merged into branches:
- 1.10: http://svn.apache.org/r1860212
- 1.8: http://svn.apache.org/r1860222
- 1.6: http://svn.apache.org/r1860242
- 1.4: http://svn.apache.org/r1860251 (test ignored)
- 1.2: http://svn.apache.org/r1860252 (test ignored)
- 1.0: http://svn.apache.org/r1860256 (test ignored)
was (Author: mreutegg):
Merged into branches:
- 1.10: http://svn.apache.org/r1860212
- 1.8: http://svn.apache.org/r1860222
- 1.6: http://svn.apache.org/r1860242
> Additional test for OAK-8012
> ----------------------------
>
> Key: OAK-8353
> URL: https://issues.apache.org/jira/browse/OAK-8353
> Project: Jackrabbit Oak
> Issue Type: Test
> Components: documentmk
> Reporter: Marcel Reutegger
> Assignee: Marcel Reutegger
> Priority: Minor
> Fix For: 1.10.3, 1.14.0, 1.6.18, 1.8.14
>
>
> It turns out the issue described in OAK-8012 may also happen without a
> restart. An incorrect read may happen when the orphaned branch clean up kicks
> in. There is a race condition in the code before OAK-8012 changes. The read
> may check the state of an orphaned branch and doesn't see it in the
> {{UnmergedBranches}} map while it is still visible in the DocumentStore.
> This issue is about an additional test to verify OAK-8012 also fixes this
> situation.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
