Hi,
I don't understand your first point: even if the bundle is installed, it
doesn't mean the datasource is there. If you remove the blueprint
datasource descriptor from the deploy folder, the datasource is removed.
The bundle stays there, but it's not a problem.
More over, as jdbc:create can install the "driver" bundle, I can add an
option to uninstall the "driver" bundle.
Regards
JB
On 11/02/2014 08:05 PM, boodoopl wrote:
Hi,
I've tested both:
* With derby-datasoruce bundle: it works fine, but it is against code
reusability rule:P The datasource cannot be uninstalled by jdbc:delete
command, because the bundle is not inside deploy directory, so there is some
exception about that
* With jdb-default-selector bundle: I was not able to make it working on
initial custom Karaf distribution startup - I mean datasource was not ready
and I got some exceptions from JPA - but when I started Karaf second time
everything worked fine; the same situation during integration tests - they
also failed due to missing datasource
So I also vote for 1.
But I found third possibility introduced by Karaf 3.0.2 KARAF-3034
<https://issues.apache.org/jira/browse/KARAF-3034>
If I understand the feature correctly: If a special .etc file is put inside
the custom distribution the datasource will be created automatically.
What is unclear for me:
* What should be the etc file name
* What should be the etc file content
* How to add etc file to custom distrubution by karaf-maven-plugin
--
View this message in context:
http://karaf.922171.n3.nabble.com/Creating-Karaf-custom-distribution-with-JDBC-data-source-tp4036058p4036195.html
Sent from the Karaf - User mailing list archive at Nabble.com.
--
Jean-Baptiste Onofré
[email protected]
http://blog.nanthrax.net
Talend - http://www.talend.com
