Well, at the end I figured it out...
Rickard posted its solution and I implemented it;
Rickard restored the throw new RemoteException("Reentrant call") (and so I
did) that is now remarked (around line 185 in EntityInstanceInterceptor).
This led to the break of my 11-th test, 'cause when I remarked it locktest
run fine.
This part of code is entered because the tx associated with the context is
null.
Soo, as it is now in CVS (not spec compliant, but works) + Rickard's
solution works <happy>fine</happy>.
Bye
Simon
> -----Original Message-----
> From: Bordet, Simone
> Sent: gioved� 19 ottobre 2000 16:27
> To: 'jBoss Developer'
> Subject: RE: [jBoss-Dev] Release lock bug in JDK? Fixed ? :-(
>
>
> Baahhh,
>
> too early. I run 10 test successfully and I said "it works",
> guess what
> happened to the 11-th test ?
>
> Simon
>
> > -----Original Message-----
> > From: Bordet, Simone
> > Sent: gioved� 19 ottobre 2000 15:58
> > To: 'jBoss Developer'
> > Subject: RE: [jBoss-Dev] Release lock bug in JDK? Fixed ! :-)
> >
> >
> > Hey Rickard,
> >
> > > > > and just lock on the context since at least it seems
> > > > > that it does
> > > > > release locks on the object it has.
> > > > > ideas? (btw the test code is in jbosstest and is called
> > > "locktest"
> > > >
> > > > I'd suggest to use the instance cache object also for wait
> > > and notifyAll
> > > > instead of the context. Since the logic for wait and
> > > notifyAll is in the
> > > > int'rs and they all can access the container, they all can
> > > access the
> > > > instance cache.
> > >
> > > Works but would be too slow. Too many threads waking up to
> > > see "oh, it was
> > > not for me". Nah, the semaphore stuff I submitted should
> > work better.
> >
> > You're right, as usual !
> >
> > I implemented the solution you suggested and it works smoothly :-))
> >
> > Marc, Sebastien, any result / comment from you ?
> >
> > BTW, I've also implemented the EJX part for the new caches,
> > and I'm ready to
> > commit it. With this it comes also the standardjboss.xml
> > where I put as
> > standard cache policy the NoPassivationCachePolicy. Is this
> > OK for you ?
> >
> > Regards,
> >
> > Simon
> >
>
