jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Fabián Mandelbaum <fmandelb...@gmail.com>
Subject Re: non-versioned "cruft"
Date Tue, 24 May 2011 10:33:53 GMT
Hello John. From time to time you have to run a process analog to
memory garbage collection, on the repository. I had the same question
a few weeks ago on this list. You can take a look here:

http://wiki.apache.org/jackrabbit/DataStore#Data_Store_Garbage_Collection

Good luck.

On Mon, May 23, 2011 at 5:37 PM, John Langley <langleyatwork@gmail.com> wrote:
> We are using a jackrabbit 2.2.5 installation to store both versioned and
> non-versioned files and our only production interface is via webdav.
>
> Over time we've noticed that when we migrate a repository using the
> RepositoryCopier tool that the size of the stored data drops dramatically.
> In one instance it was 1/35th of the size of the un-migrated data set as
> measured by doing a mysql dump of the database. Part of our migration
> process is to run a comparison tool comparing the old and new repositories.
> Using this interface we check every file including the versioned files so we
> know that we've had a successful copy.
>
> Consequently, our conclusion is that this "cruft" is comprised of "orphaned"
> nodes that are associated with non-versioned content. Does this make sense?
> If so, is there a suggested way to prune out this unused content?
>
> Thanks in advance,
>
> -- Langley
>



-- 
Fabián Mandelbaum
IS Engineer

Mime
View raw message