Hi all,

Please pardon me, my ODMG ignorance is shining through ;)

Mark

Thomas Mahler wrote:
> 
> Hi again,
> 
> Anil is right. But the Isolationlevel stuff does only work inside the
> ODMG layer. It is not implemented on the PB level!
> 
> The isolation-levels in the repository have nothing to do with isolation
> levels on the database levels. They are not translated into JDBC calls.
> They just specify the behaviour of the OJB/ODMG Lockmanager.
> 
> cheers,
> Thomas
> 
> Mark Howell wrote:
> > Anil,
> >
> > I think that attribute option has to do with database interaction, and
> > not with how the PB and it's cache materialize and distribute objects.
> > I'm not sure about this, but I've tested several of these modes
> > including serializable, repeatable-read and read-uncommitted and none of
> > those modes made the PB return distinct instances of an entity to
> > multiple threads.  My tests were only with the PB API.
> >
> > Thanks,
> >
> > Mark
> >
> >
> > Anil Sharma wrote:
> >
> >>Wouldn't specifying the transaction isolation level to
> >>"RepeatableReads/Serializable" for the class(es) in question achieve that ?
> >>
> >>Here's an excrept from lockmanager documentation ::
> >>
> >>/////
> >>The ODMG specification does not prescribe transaction isolationlevels or
> >>locking strategies to be used. Thus there are no API calls for setting
> >>isolationlevels. OJB provides four different isolationlevels that can be
> >>configured for each persistent class in the XML repository.
> >>The isolationlevel of a class can be configured with the following attribute
> >>to a ClassDescriptor:
> >>
> >><ClassDescriptor id="1" isolation="read-uncomitted">
> >>///////
> >>
> >>I am using PB Query objects (PB API) within ODMG transactions (ODMG API) -
> >>so I believe you can use both of them on the same repository/db.
> >>
> >>HTH
> >>
> >>Anil
> >>
> >>-----Original Message-----
> >>From: Mark Howell [mailto:[EMAIL PROTECTED]]
> >>Sent: Monday, October 14, 2002 5:38 PM
> >>To: OJB Users List
> >>Subject: Re: objects
> >>
> >>Hello,
> >>
> >>Is there a way to force the PB to *not* share local instances of an
> >>Object that it finds with multiple threads?  I would prefer, for at
> >>least some of my classes, that one thread not see the changes made by
> >>another unless that object was specifically stored in the PB.
> >>
> >>Would I need to use the ODMG API to do this currently?  If so, can I use
> >>the PB API and ODMG concurrently on the same repository/db?
> >>
> >>Thanks,
> >>Mark
> >>
> >>Thomas Mahler wrote:
> >>
> >>>Kevin Viet wrote:
> >>>
> >>>>If two threads query the API for this object, they share the same
> >>>>reference on this object.
> >>>
> >>>correct.
> >>>
> >>>
> >>>>So if a thread change a value (without update the object through the
> >>>>API) of the object, the other thread will see this change (even for the
> >>>>object cache)
> >>>
> >>>correct for PB API. For ODMG we have a pessimistic locking that will
> >>>protect against such conflicts.
> >>>
> >>>
> >>>>Won't any problem arise if thread1 change some values and thread2 is
> >>>>commiting the update.
> >>>
> >>>correct
> >>>
> >>
> >>--
> >>To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
> >>For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
> >>
> >>--
> >>To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
> >>For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
> >
> >
> > --
> > To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
> > For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
> >
> >
> >
> >
> 
> --
> To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>

--
To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>

Reply via email to