jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Felix Meschberger (JIRA)" <j...@apache.org>
Subject [jira] Commented: (JCR-20) Logging into a repository with a big version history takes a long time
Date Thu, 18 Nov 2004 16:28:35 GMT
     [ http://nagoya.apache.org/jira/browse/JCR-20?page=comments#action_55618 ]
Felix Meschberger commented on JCR-20:

Raising the Java VM max heap helps in this case. But given that jackrabbit uses around 60MB
(or more) when logging in, this kind of worries me...

> Logging into a repository with a big version history takes a long time
> ----------------------------------------------------------------------
>          Key: JCR-20
>          URL: http://nagoya.apache.org/jira/browse/JCR-20
>      Project: Jackrabbit
>         Type: Bug
>  Environment: Jackrabbit SVN 76106
>     Reporter: Felix Meschberger
>     Priority: Critical

> Wenn a SessionImpl instance is created, the VersionManager.getVirtualItemStateProvider
method is called. This method - amongst other things - loads the complete (!) version history
into memory and walks through it to do some mapping.
> Besides taking a long time (near 1 minute just to get the version history through PersistentVersionManager.getVersionHistories())
mapping the version histories ultimately results in an "OutOfMemoryError".
> Currently there are 768 version histories and this is only a very small fraction of the
expected final number of version histories in my application

This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
If you want more information on JIRA, or have a bug to report see:

View raw message