Wouldn't it be more appropriate to configure the second repository as
a clustered (http://wiki.apache.org/jackrabbit/Clustering) instance?
I think that Phillip wants to have the second repository's indexed
updated when there's any data change.
Alessandro
On Jul 27, 2007, at 9:52 AM, Stefan Guggisberg wrote:
hi philip,
On 7/26/07, Phillip Rhodes <[EMAIL PROTECTED]> wrote:
Hello,
I have a instance of jackrabbit running with
OraclePersistenceManager on a remote machine.
I would like to setup a second (readonly) repository on my local
machine that shares the same data as the aforementioned repository
so that I can do some analysis on some performance problems that I
am having.
I thought by just starting up the repository on my local machine
with the OraclePersistenceManager configured the same as the first
repository (same database instance) would cause jackrabbit to
recreate the workspace directories and index directories and set
them up for use, but alas, this is not the case.
Any xpath query that I perform fails to return anything.
I would appreciate any help and guidance on getting past this so I
can dig deeper in what part of my usage of jackrabbit is causing
problems. Is there a way to instruct jackrabbit to refresh the
filesystem using the PersistenceManager as it's source for data?
please note that the repository home directory contains vital data
(such as custom node types and namespaces). you need to copy the
contents of your repository home directory
(except the index folders) to the other machine.
alternatively you could configure jackrabbit to store everything
(except the index) in the db
by using DbFileSystem instead of LocalFileSystem. however, i
wouldn't recommend
it in general as you cannot easily read/edit the information stored in
the DbFileSystem.
btw: here's a related issue: http://issues.apache.org/jira/browse/
JCR-964
cheers
stefan
Thanks.
Phillip
