jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Marcel Reutegger (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (JCR-3658) MembershipCache not consistently synchronized
Date Thu, 29 Aug 2013 15:29:53 GMT

    [ https://issues.apache.org/jira/browse/JCR-3658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13753725#comment-13753725
] 

Marcel Reutegger commented on JCR-3658:
---------------------------------------

The test creates 100 users and 8 groups. Each user belongs to those 8 groups. Then 8 reader
sessions are created and they start calling User.memberOf(), which in turn will use the MembershipCache.
To simulate cache misses, the readers are ramped up with more users over time and they randomly
pick from the known users. While readers use the cache the main thread will write to the repository
to show the effect on the cache usage.

There's also a test method, which runs the test many times to get a more consistent view of
the timing.
                
> MembershipCache not consistently synchronized
> ---------------------------------------------
>
>                 Key: JCR-3658
>                 URL: https://issues.apache.org/jira/browse/JCR-3658
>             Project: Jackrabbit Content Repository
>          Issue Type: Bug
>          Components: jackrabbit-core, security
>    Affects Versions: 2.7
>            Reporter: Tobias Bocanegra
>            Assignee: Marcel Reutegger
>            Priority: Minor
>         Attachments: JCR-3658-test.patch
>
>
> the membership cache access is mostly synchronized on 'this' but in the onEvent() handler,
the internal cache object is synchronized.
> suggest: to improve cache access by a r/w lock instead.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message