jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Paco Avila <monk...@gmail.com>
Subject Re: alternative export - import strategy
Date Tue, 03 Feb 2009 21:47:51 GMT
We application is developed to run on JBoss. Default repository.xml is
used, so all files and documents are stored in server filesystem. To
make a backup, I stop JBoss and copy the repository folder to another
location using rsync to minimize copy time. After that start JBoss
again. This work is done using a bash script.

I have seen that jcr-backups work only for Jackrabbit 1.3 so I wonder
if this utility can manage 1.4 or 1.5 repositories. I plan to test it
someday :)

On Tue, Feb 3, 2009 at 7:58 PM, Shailesh Mangal
<shailesh.mangal@getzephyr.com> wrote:
> While database export and import is not working for us when we move from
> windows to linux, I would like to explore alternative export - import
> strategies e.g. http://sourceforge.net/projects/jcr-backup
>
> We have lots of binary data, files etc. I would like to know if someone has
> successfully implemented backup and restore strategy and the tools used to
> accomplish the same.
>
> -Shailesh
>
> On Feb 3, 2009, at 10:21 AM, Shailesh Mangal wrote:
>
>>
>> On Feb 3, 2009, at 10:08 AM, Alexander Klimetschek wrote:
>>
>>> On Tue, Feb 3, 2009 at 6:36 PM, Shailesh Mangal
>>> <shailesh.mangal@getzephyr.com> wrote:
>>>>
>>>> I noticed that everything (including workspace.xml) gets recreated (on
>>>> mac
>>>> and windows). I am witnessing this issue on Linux (Fedora) that the
>>>> indexing
>>>> are not getting rebuilt.
>>>
>>> Are you sure workspace.xml gets re-created properly (ie. compare old
>>> and new workspace.xml)?
>>
>> Yes, we only have default workplace.
>>>
>>>
>>> I assume you only have the "default" workspace that gets automatically
>>> created on startup and copies the template values from the workspace
>>> section in repository.xml, that are probably the same for your only
>>> workspace, and once the workspace's persistence manager starts and
>>> connects to the DB, it (coincidentally) detects the data that is
>>> already existing. This approach would definitely fail as soon as you
>>> would have two slightly differently configured workspaces...
>>>
>>>> BTW, I have also tried deleting just the indexes, same behavior.
>>>
>>> Can you access the nodes and properties, are they present? The index
>>> is not required for navigational access, only for search operations.
>>> And do you see anything in the logs?
>>
>> Nothing in logs. Can't access nodes and properties. My datafile export
>> size is 105MB and data shows up fine in xp.
>>
>>>
>>>
>>> Regards,
>>> Alex
>>>
>>> --
>>> Alexander Klimetschek
>>> alexander.klimetschek@day.com
>
>
>
>



-- 
Paco Avila
GIT Consultors
tel: +34 971 498310
fax: +34 971496189
e-mail: pavila@git.es
http://www.git.es

Mime
View raw message