The root issue is the infinispan tests afaict. They seem to be trying to
create these objects without passing in the Settings, which is how
HibernateTransactionManagerLookup is set up to locate the JtaPlatform to use
to obtain the TransactionManager. I'll fix that too
On Monday, March 07, 2011, at 03:59 am, Galder Zamarreño wrote:
Steve,
The root cause appears to be
http://opensource.atlassian.com/projects/hibernate/browse/HHH-5949
Since then, HibernateTransactionManagerLookup does not return a transaction
manager when requested one.
As a result of this, the following set up code has changed in
CacheTestUtil:
cfg.setProperty( JtaPlatformInitiator.JTA_PLATFORM,
BatchModeJtaPlatform.class.getName() );
Either BatchModeJtaPlatform or the some code around it is not returning a
transaction manager when it should.
Tracing the code, I can see that when
HibernateTransactionManagerLookup.getTransactionManager() calls
jtaPlatform.retrieveTransactionManager(), then jtaPlattform is pointing
to: org.hibernate.service.jta.platform.spi.JtaPlatform_$$_Proxy@1283730842
And the call to retrieveTransactionManager does not land on
BatchModeJtaPlatform.retrieveTransactionManager()
That's the reason why no transaction manager is associated with Infinispan.
I dunno the root cause though.
Cheers,
On Mar 7, 2011, at 10:32 AM, Galder Zamarreño wrote:
> Steve,
>
> I've replicated the failures on my env. It would appear as if the
> Infinispan layer is not being configured with a transaction manager any
> more in master.
>
> I'm trying to dig further and will provide more details later on.
>
> Cheers,
>
> On Mar 4, 2011, at 4:37 PM, Galder Zamarreño wrote:
>> Ok. I'll have a look to it on Monday.
>>
>> On Mar 4, 2011, at 3:53 PM, Steve Ebersole wrote:
>>> Just on master. I have not verified 3.6. There were quite a few
>>> failures prior to my recent merge. I think my recent merge introduced
>>> a few more. I just need help tracking down what the cause of these
>>> failures are.
>>>
>>> On Friday, March 04, 2011, at 04:13 am, Galder Zamarreño wrote:
>>>> Btw, just checked 3.6.x and it looks clean, so maybe something you
>>>> guys changed in master?
>>>>
>>>> On Mar 4, 2011, at 11:12 AM, Galder Zamarreño wrote:
>>>>> In 3.6.x or master? Or both?
>>>>>
>>>>> On Mar 3, 2011, at 9:27 PM, Steve Ebersole wrote:
>>>>>> There are currently a bunch of test failures in
>>>>>> hibernate-infinispan. Are these expected? I thought they were
>>>>>> related to my changes, but i verified that these failures are
there
>>>>>> even on master.
>>>>>>
>>>>>>
>>>>>> ---
>>>>>> Steve Ebersole <steve(a)hibernate.org>
>>>>>>
http://hibernate.org
>>>>>> _______________________________________________
>>>>>> hibernate-dev mailing list
>>>>>> hibernate-dev(a)lists.jboss.org
>>>>>>
https://lists.jboss.org/mailman/listinfo/hibernate-dev
>>>>>
>>>>> --
>>>>> Galder Zamarreño
>>>>> Sr. Software Engineer
>>>>> Infinispan, JBoss Cache
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> hibernate-dev mailing list
>>>>> hibernate-dev(a)lists.jboss.org
>>>>>
https://lists.jboss.org/mailman/listinfo/hibernate-dev
>>>>
>>>> --
>>>> Galder Zamarreño
>>>> Sr. Software Engineer
>>>> Infinispan, JBoss Cache
>>>
>>> ---
>>> Steve Ebersole <steve(a)hibernate.org>
>>>
http://hibernate.org
>>
>> --
>> Galder Zamarreño
>> Sr. Software Engineer
>> Infinispan, JBoss Cache
>>
>>
>> _______________________________________________
>> hibernate-dev mailing list
>> hibernate-dev(a)lists.jboss.org
>>
https://lists.jboss.org/mailman/listinfo/hibernate-dev
>
> --
> Galder Zamarreño
> Sr. Software Engineer
> Infinispan, JBoss Cache
>
>
> _______________________________________________
> hibernate-dev mailing list
> hibernate-dev(a)lists.jboss.org
>
https://lists.jboss.org/mailman/listinfo/hibernate-dev
--
Galder Zamarreño
Sr. Software Engineer
Infinispan, JBoss Cache