Hi,
we use following two calls to shut it down
// Release all persistence broker instances
PersistenceBrokerFactory.releaseAllInstances();
// Release all pooled db connections
ConnectionFactoryFactory.getInstance().createConnectionFactory().releaseAllResources();
however it could be that
Greetings all,
first off - all the different hints of how not to leave open
connections at webapp shutdown (ie using DBCP in Tomcat and supply JNDI
DataSource to OJB, or explicitly calling release methods in OJB at
shutdown) are equally good ways of doing.
Danilo Tommasina wrote:
however it
Thanks for all your very useful replys.
We do all closing of resources in the destroy() method of our setup servlet.
We use the ConnectionFactoryPooledImpl so releasing of the resources used
should work fine.
I would have like to have used the ConnectionFactoryDBCPImpl as that is what
is used
Is my assumption wrong? In this case:
reference-descriptor name=clientRelationship
class-ref=model.client.ClientRelationship
foreignkey field-ref=objectId/
/reference-descriptor
Isn't objectId supposed to be the primary key field of
model.client.ClientRelationship?
Hi Wes,
Is my assumption wrong? In this case:
reference-descriptor name=clientRelationship
class-ref=model.client.ClientRelationship
foreignkey field-ref=objectId/
/reference-descriptor
Isn't objectId supposed to be the primary key field of
model.client.ClientRelationship?
No. OJB currently only joins on the related object primary key (which
it already knows from the mapping), so what is needed is the foreign key
field in NewClientGroup that relates to ClientRelationship.
Remember that in Reference or Collection Descriptors it is always
looking for the foreign
No problem. I think this is everything needed:
package model;
// All persistent Object extend this class, the unique object ID
// is from this class along with some other fields
public class AdmObject implements Serializable{
private int objectId = -1;
// A few other
Hi Wes,
comments see below
Lemke, Wesley wrote:
No problem. I think this is everything needed:
package model;
// All persistent Object extend this class, the unique object ID
// is from this class along with some other fields
public class AdmObject implements Serializable{
I added the (int) clientRelationshipId to NewClientGroup and removed the
access=anonymous from the field-descriptor. I am still getting the
same error.
-Original Message-
From: Armin Waibel [mailto:[EMAIL PROTECTED]
Sent: Thursday, March 24, 2005 9:04 AM
To: OJB Users List
Subject: Re:
Nathan Smith wrote:
I cannot quite remember what the problem was, but it had something to with
Iterators not returning all objects or Collections being returned containing
no objects, but thats better left for another time at the moment.
Aha, this smells like objects of class OJBIterator returned
Lemke, Wesley wrote:
I added the (int) clientRelationshipId to NewClientGroup and removed the
access=anonymous from the field-descriptor. I am still getting the
same error.
Damned! What a nasty problem ;-)
The error message seems to be clear
Error 500: Server caught unhandled exception from
forget my previous post, the error message isn't clear. Why does OJB
lookup 'clientRelationshipId' in class 'ClientRelationship' instead of
'NewClientGroup'?
Could you post the whole stack trace?
Armin
Armin Waibel wrote:
Lemke, Wesley wrote:
I added the (int) clientRelationshipId to
I have a ParentObject which owns a ChildObject. The ChildObject is not
null, but contains all null fields. The constructor of the ParentObject
creates a ChildObject which is not null, but contains all null fields.
When OJB loads ParentObject, it calls the default constructor which
creates a
Hehe, that was my originally question when I changed it from objectId to
clientRelationshipId :)
Here is the entire stack trace:
[3/24/05 9:08:43:844 CST] 1985c96 SystemOut O
[org.apache.ojb.broker.accesslayer.sql.SqlGeneratorDefaultImpl] DEBUG:
SQL:SELECT A0.agr_approver,A0.client_id FROM
Lemke, Wesley wrote:
Hehe, that was my originally question when I changed it from objectId to
clientRelationshipId :)
I run a test case with same mapping (1:n with 1:1 back-reference and
shared FK) with latest from OJB_1_0_RELEASE branch. The test pass
without problems.
Maybe the problem is the
That worked Armin, thanks!
One question though, those mappings were correct, and should work,
right? Will it hurt anything to not have them in there?
I created a Junit Test. It works with those mappings removed, but fails
when the lines are there. Would you like me to send it to you?
Lemke, Wesley wrote:
That worked Armin, thanks!
puh! This is a good news.
One question though, those mappings were correct, and should work,
right?
In impulse, yep it should work. But there can be a problem with
inheritance and the use of SequenceManagerNativeImpl, because this SM is
not
Just checking if anyone knows the target date for 1.02? I am looking to
upgrade but am afraid of the cache problem in 1.01.
Ryan
Ryan Vanderwerf wrote:
Just checking if anyone knows the target date for 1.02? I am looking to
upgrade but am afraid of the cache problem in 1.01.
Last mail from Brian announce the release for today or tomorrow (when he
has a change to build the release).
regards,
Armin
Ryan
Brian is working on the release as we speak.
R
On Mar 24, 2005, at 12:08 PM, Ryan Vanderwerf wrote:
Just checking if anyone knows the target date for 1.02? I am
looking to
upgrade but am afraid of the cache problem in 1.01.
Ryan
!DSPAM:42434831247681776579306!
Robert S. Sfeir
Am pushing it tonight. Won't announce until tomorrow night though so
that it can mirror out first =)
-Brian
On Mar 24, 2005, at 12:08 PM, Ryan Vanderwerf wrote:
Just checking if anyone knows the target date for 1.02? I am looking to
upgrade but am afraid of the cache problem in 1.01.
Ryan
21 matches
Mail list logo
