On Dec 15, 2005, at 11:42 AM, (Robert Van Overmeiren) wrote:
I thought the export/import was more for migrating config trees
from one
release to another rather than backup, but I see how being able to do
this with the content is a good thing.
It's a lot like the typical PostgreSQL or MySQL backup routines where
you dump everything to SQL. From both a system administrator and
developer's stand point, I like having quick access to things like a
database or JCR dump that are simple to understand and restore.
I've dabbled with using ANT to zip up backups, but since it was
slow and
sometimes ran out of memory was advised to use shell scripting.
That was
a little slow too until I switched to having it gzip the files. I have
it create separate gzip files for public docroot, repositories, &
templates, and author docroot, repositories, & templates. Should a
problem arise, I suspect it would be in the repository, so would be
able
to restore just that. Have never had to though.
Now that persistence is based on Berkley DB, maybe this wouldn't work
anymore.
With regards to the repositories I think this approach is still fine
w/Berkeley DB and is certainly a good idea for Magnolia production
environments:
http://www.sleepycat.com/jedocs/GettingStartedGuide/backup.html
- Tom
----------------------------------------------------------------
for list details see
http://www.magnolia.info/en/magnolia/developer.html
----------------------------------------------------------------
