This is wrong. You shouldn't be sharing a read/write DataContext between
requests. I can't say with 100% certainty that this is causing this particular
problem, but it is guaranteed to cause various consistency issues.
I often use a pattern of 1 shared app-scoped DataContext for reads, and then
create short-lived throwaway DataContexts to change objects ("short-lived"
means a lifespan <= request duration).
Andrus
On Jan 24, 2011, at 12:35 PM, Bruno René Santos wrote:
> In the init() phase of each portlet (I'm using liferay) I call the
> DataContext.createDataContext() to create a DataContext that is used
> throughout the application (always the same instance, I create an object that
> is passed as parameter to all other objects). And yes my app is not readonly
> but I also do not try to write on the view... Have you ever thought how we
> could handle these cases near the comment you left on the code? Could I try
> like a commitChanges there? Any ideas are welcome :)
>
> Thanx
> Bruno
>
> Em 24-01-2011 10:18, Andrus Adamchik escreveu:
>>> I am developing a web application so it is possible that some thread is
>>> interfering on the refreshing of the Object Cache?
>> This depends on how you share a DataContext. If your app is not read-only,
>> DataContext should not be shared.
>>
>> Andrus
>>
>> On Jan 24, 2011, at 12:09 PM, Bruno René Santos wrote:
>>
>>> HoVPlanoContas is a view. And yes the id 129 exists. On that piece of code
>>> you're expecting a committed object but you're getting a modified one... I
>>> am going to analyze a little further why on this region of code this
>>> happens (On other parts I dont get this error). But it would help to know
>>> the purpose of this PrepareForAccess method and how the object starts as
>>> HOLLOW and all of a sudden it is supposed to be committed (but in the end
>>> is modified...). I am developing a web application so it is possible that
>>> some thread is interfering on the refreshing of the Object Cache? Any other
>>> ideas?
>>>
>>> Thanx
>>> Bruno
>>>
>>> Em 24-01-2011 09:43, Andrus Adamchik escreveu:
>>>> Can you confirm that there is a record for HoVPlanoContas in the database
>>>> with id of 129? Is it reproducible?
>>>>
>>>> Andrus
>>>>
>>>> On Jan 24, 2011, at 3:03 AM, Bruno René Santos wrote:
>>>>
>>>>> Hello all,
>>>>>
>>>>> Im having an error when I access an object:
>>>>>
>>>>> Caused by: org.apache.cayenne.FaultFailureException: [v.3.0.1 Aug 25 2010
>>>>> 19:38:17] Error resolving fault for ObjectId:<ObjectId:HoVPlanoContas,
>>>>> id=129> and state (modified). Possible cause - matching row is missing
>>>>> from the database.
>>>>> at
>>>>> org.apache.cayenne.BaseContext.prepareForAccess(BaseContext.java:171)
>>>>> at
>>>>> org.apache.cayenne.CayenneDataObject.readProperty(CayenneDataObject.java:230)
>>>>>
>>>>> When I go to the line 171 of BaseContext there is this comment from
>>>>> Andrus:
>>>>>
>>>>> // TODO: andrus 4/13/2006, modified and deleted states are possible due to
>>>>> // a race condition, should we handle them here?
>>>>>
>>>>> Why this happens? Any solutions?
>>>>>
>>>>> Thanx
>>>>> Bruno
>>>>> --
>>>>> Bruno René Santos | [email protected]<mailto:[email protected]> |
>>>>> Gestor de Projectos | Analista | Programador | Investigador
>>>>>
>>>>> Holos - Soluções Avançadas em Tecnologias de Informação S.A.
>>>>> Parque de Ciência e Tecnologia de Almada/Setúbal . Edifício Madan Parque
>>>>> Rua dos Inventores . Quinta da Torre . 2825 - 182 Caparica . Portugal
>>>>> Phone: +351 210 438 686 . Fax: +351 210 438 687 . Web: www.holos.pt
>>>>>
>>>>>
>>>>> This email and any files transmitted with it are confidential and
>>>>> intended solely for the use of the individual or entity to whom they are
>>>>> addressed. If you are not the intended recipient or the person
>>>>> responsible for delivering the email to the intended recipient, be
>>>>> advised that you have received this email in error and that any use,
>>>>> dissemination, forwarding, printing, or copying of this email is strictly
>>>>> prohibited. If you have received this email in error please notify Bruno
>>>>> René Santos by telephone on +351 210 438 686
>>>>>
>>>>
>>>
>>> --
>>> Bruno René Santos | [email protected]<mailto:[email protected]> | Gestor
>>> de Projectos | Analista | Programador | Investigador
>>>
>>> Holos - Soluções Avançadas em Tecnologias de Informação S.A.
>>> Parque de Ciência e Tecnologia de Almada/Setúbal . Edifício Madan Parque
>>> Rua dos Inventores . Quinta da Torre . 2825 - 182 Caparica . Portugal
>>> Phone: +351 210 438 686 . Fax: +351 210 438 687 . Web: www.holos.pt
>>>
>>>
>>> This email and any files transmitted with it are confidential and intended
>>> solely for the use of the individual or entity to whom they are addressed.
>>> If you are not the intended recipient or the person responsible for
>>> delivering the email to the intended recipient, be advised that you have
>>> received this email in error and that any use, dissemination, forwarding,
>>> printing, or copying of this email is strictly prohibited. If you have
>>> received this email in error please notify Bruno René Santos by telephone
>>> on +351 210 438 686
>>>
>>
>>
>
>
> --
> Bruno René Santos | [email protected] <mailto:[email protected]> | Gestor
> de Projectos | Analista | Programador | Investigador
>
> Holos - Soluções Avançadas em Tecnologias de Informação S.A.
> Parque de Ciência e Tecnologia de Almada/Setúbal . Edifício Madan Parque
> Rua dos Inventores . Quinta da Torre . 2825 - 182 Caparica . Portugal
> Phone: +351 210 438 686 . Fax: +351 210 438 687 . Web: www.holos.pt
>
>
> This email and any files transmitted with it are confidential and intended
> solely for the use of the individual or entity to whom they are addressed. If
> you are not the intended recipient or the person responsible for delivering
> the email to the intended recipient, be advised that you have received this
> email in error and that any use, dissemination, forwarding, printing, or
> copying of this email is strictly prohibited. If you have received this email
> in error please notify Bruno René Santos by telephone on +351 210 438 686
>
