[
https://issues.apache.org/jira/browse/OAK-3924?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15119323#comment-15119323
]
Tomek Rękawek edited comment on OAK-3924 at 1/27/16 2:42 PM:
-------------------------------------------------------------
OK, so we need to do following things here:
1. remove the Oak-level locking from the patch, as it doesn't fix the issue
anyway,
2. sort updates before sending them to DB,
3. disable the bulk updates by default and enable them with a system property.
[~reschke], does it make sense to you?
was (Author: tomek.rekawek):
OK, so we need to do following things here:
1. remove the Oak-level locking from the patch, as it doesn't fix the issue
anyway,
2. sort updates before sending them to DB,
3. disable the bulk updates by default and enable them with a system property.
Makes sense?
> Fix database-level row deadlock during bulk updates in RDB
> ----------------------------------------------------------
>
> Key: OAK-3924
> URL: https://issues.apache.org/jira/browse/OAK-3924
> Project: Jackrabbit Oak
> Issue Type: Sub-task
> Components: rdbmk
> Reporter: Tomek Rękawek
> Priority: Critical
> Fix For: 1.4
>
> Attachments: OAK-3924.patch
>
>
> It seems that the new bulk createOrUpdate() implementation in RDB is prone
> for the deadlocks. It isn't a bug in the Oak code, but rather something
> related to the database implementations. The bug can be observed if we have
> multiple simultaneous bulk updates and some of the rows repeat among them.
> The attached patch contains an unit test {{testConcurrentWithConflict}}
> presenting the issue.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
