Re: [hibernate-dev] JPA2 locking
On Tue, 2009-10-20 at 14:12 -0400, Scott Marlow wrote:
On 10/20/2009 12:16 PM, Steve Ebersole wrote:
> See my reading of P2 though is more than just ensuring a repeatable
> read.
>
> If T1 does lock(entity, LockModeType.PESSIMISTIC_READ) we need to make
> sure T2 cannot update or delete that row b4 T1 completes. That's
> blocking, not simply repeatable read.
>
I guess we get into how repeatable read isolation level is implemented
and whether it could be implemented without at least a read lock on the
database table row. Or did you mean something else?
That's true. And actually
the dialects have a method to help with this:
Dialect#doesRepeatableReadCauseReadersToBlockWriters :)
There is also, Dialect#doesReadCommittedCauseWritersToBlockReaders
--
Steve Ebersole <steve(a)hibernate.org>
Hibernate.org