What do you need this class for?
Does the conflict actually has sth todo with the Exception or do you need
this Class ** for the Spring <=> OpenJPA integration?
Cause I don't see any reference to this class if you follow the Spring/JPA
Guide:
http://static.springsource.org/spring/docs/3.0.x/reference/orm.html#orm-jpa

You could add the quick & dirty solution of Vasiliy anyway, if it really
improves the situation.

Sebastian

2011/8/9 Maxim Solodovnik <[email protected]>

> Hello Sebastian,
>
> I was unable to take a look at your latest logs, but suppose it is also
> "InvalidStateException: This operation cannot be performed while a
> Transaction is active."
>
> To handle it I add EntityManager injection via Spring config as was
> previously discussed.
>
> Unfortunately I'm currently stuck on following library conflict:
> 1) red5 is dependant on javaee-api-5.1.1.jar which
> contains javax\persistence\spi\PersistenceProvider definition
> 2) and openjpa-all-2.1.0.jar contains definition of this class
> as well as implementation
> (org.apache.openjpa.persistence.PersistenceProviderImpl)
>
> The problem is: these two classes seems to be incompatible:
>
> javax.persistence.PersistenceException: The instance of the object with the
> class name 'org.apache.openjpa.persistence.PersistenceProviderImpl' in the
> ClassLoader 'WebappClassLoader' is not an instance of PersistenceProvider
> interface.
>
> I tried to get fresh version of javaee-api*.jar
>
> but
> 1) the version downloaded from maven repository javaee-api-6.0.jar doesn't
> have all necessary packages
> 2) the version downloaded from
> http://www.jarvana.com/jarvana/browse/org/ow2/jonas/osgi/javaee-api/ has
> all packages BUT has incompatible version of PersistenceProvider interface
>
> Maybe you have idea how can I resolve this?
> Thanks a lot in advance
>
> p.s. May be I can implement DAOTransaction abstract class (from initial
> Vasiliy's proposal) as quick and dirty solution?
>
>
> On Mon, Aug 8, 2011 at 15:58, [email protected] <[email protected]
> > wrote:
>
>> meanwhile the service become unavailable,
>> you might grab the latest Logfile:
>> http://demo.openmeetings.de/jvm.stdout
>>
>> I might switch back to the previous version later today.
>>
>> We might at least agree on a road to take to solve those issues.
>>
>>
>> Sebastian
>>
>>
>> 2011/8/8 Maxim Solodovnik <[email protected]>
>>
>>> Hello Alexei,
>>>
>>> should i take a look at all these issues? Vasiliy is on vacation right
>>> now.
>>>
>>>
>>> On Mon, Aug 8, 2011 at 00:43, [email protected] <
>>> [email protected]> wrote:
>>>
>>>> Hi,
>>>>
>>>> there are a number of Issues with openJPA and the migration.
>>>> The demo server is now updated but I needed several hours to get the
>>>> basic functions running again.
>>>>
>>>> Also the shift with the user_ids in the import/export is still todo for
>>>> the profiles_$userId thing.
>>>> Somehow we could agree that we make a unique MD5 hash for each user for
>>>> his profile path.
>>>>
>>>> The user sign up process is basically fixed, however when you repress
>>>> the button to sign up again, or you just enter an existing user again, it
>>>> will show the message that you have successfully signup, instead of showing
>>>> a message that the user/login is already taken (which is what already
>>>> implemented).
>>>>
>>>> I think there is really quite a bit to test and fix, the openmeetings
>>>> service became unresponsible/did not react to any user interaction after
>>>> just 30 minutes without any message about the reason in the logfile.
>>>>
>>>> Also there are Issue with the Caching. Example:
>>>> I do create a new conference room with type restricted, max number 16,
>>>> type public
>>>> goto conference rooms, see the conference room,
>>>> go back to administration, edit that room set max number to 100,
>>>> goto conference rooms, see the conference room => STILL 16
>>>>
>>>> => That is really bad, cause to track down this problem you really will
>>>> need to rethink the hole procedure of insert/update/delete and how it
>>>> behaves by running in multiple Threads. This behaviour can lead to various
>>>> unpredictable situations, strongly depending on what usage scenario and how
>>>> many records you already have in the database
>>>> I've spend weeks to track down such problems with the previous
>>>> implementation, where the only really solution is to use the spring 
>>>> injected
>>>> session and let spring manage the session-context. However it really needs
>>>> some in depth testing. We can hardly release a package right now.
>>>>
>>>> Sebastian
>>>>
>>>>
>>>> --
>>>> Sebastian Wagner
>>>> http://www.webbase-design.de
>>>> http://openmeetings.googlecode.com
>>>> http://www.wagner-sebastian.com
>>>> [email protected]
>>>>
>>>
>>>
>>>
>>> --
>>> WBR
>>> Maxim aka solomax
>>>
>>
>>
>>
>> --
>> Sebastian Wagner
>> http://www.webbase-design.de
>> http://openmeetings.googlecode.com
>> http://www.wagner-sebastian.com
>> [email protected]
>>
>
>
>
> --
> WBR
> Maxim aka solomax
>



-- 
Sebastian Wagner
http://www.webbase-design.de
http://openmeetings.googlecode.com
http://www.wagner-sebastian.com
[email protected]

-- 
You received this message because you are subscribed to the Google Groups 
"OpenMeetings developers" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/openmeetings-dev?hl=en.

Reply via email to