Hi,
marc fleury wrote:
> this has been bothering me for quite some time...
I've been wondering too...
> did I just embarrass myself?
Not at all. Wanting to do that makes perfectly sense.
Only problem is that JTA doesn't allow it.
> do you guys know?
No, but an explanation from the JTA folks would be
nice.
Best Regards,
Ole Husgaard.
>
> marcf
>
> |-----Original Message-----
> |From: marc fleury [mailto:[EMAIL PROTECTED]]
> |Sent: Monday, July 02, 2001 6:32 PM
> |To: [EMAIL PROTECTED]
> |Subject: registerSync on rolledback transaction
> |
> |
> |Hi,
> |
> |sorry if this has been addressed before, I am a bit of a newbie on
> |this list
> |
> |I want to register a synchronization on a transaction that was
> |marked for rollback.
> |
> |I am curious as to why the spec decides to throw a RollbackException.
> |
> |I mean I *have* the tx, if I wanted to do something about a marked
> |for rollback thingy all I have to do is
> |tx.getStatus().equals(Status.blabla_marked_for_rollback) and
> |decide. If I WANT to be notified of the tx demarcation event,
> |wether that be commit or rollback, is none of the tx' business.
> |
> |In other words, I am surprised by the assymetry you artificially
> |introduce between commit and rollback on the registration of
> |synchronizations. Why should the possible outcome affect the
> |registration at all?.
> |
> |Finally, sorry if this has been adressed as well and I know this
> |is JTA, but... what is a TM to do ? :) my guts tell me "register
> |the damn synchronization" (per the above)
> |
> |please include me in any response as I am not on this list.
> |
> |thanks
> |
> |marcf
> |
> |
> |_________________
> |Marc Fleury, Ph.D
> |[EMAIL PROTECTED]
> |_________________
>
> _______________________________________________
> Jboss-development mailing list
> [EMAIL PROTECTED]
> http://lists.sourceforge.net/lists/listinfo/jboss-development
_______________________________________________
Jboss-development mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/jboss-development
