On Wed, Feb 13, 2013 at 02:00:00PM -0800, Iuri G. wrote:
>
> You are right John, there is no duplication of work. the counter is
> incremented properly. But if you look in your SQL table, you will see that
> there are couple clean_car participants in the table with expid 0_1 on all
> of them. I dont think worker picks those participants, maybe only last one,
> thats why there is no duplication, but the expression tree does not match
> to process pdef, and to some degree does not make sense to the worker.
> Maybe I am not quite groking what SQL records should look like and that's
> normal?

Hello Iuri,

as explained sooner in this thread, with a cursor/repeat, it's quite normal
to have expression with the same expid... It makes it easier to point on the
process definition tree. The subid is different.

> I just committed changes to script to match my setup.
> I ran my modified script with locking and without locking and what i found
> was that in both cases workitem counter was incremented correctly. The
> difference was that with locking, process expression looked exactly like
> PDEF in SQL table, but without locking, there were 5 extra clean_car
> participants in SQL table.

Do you mean 5 extra participants or 5 extra workitems?

Yes, according to your process definition, there should be a single workitem
alive at any time for a give workflow instance (of that definition).

So, could this whole thread be packed into:

"my custom participant implementation does not clean its workitem after it
proceeded"

?

--
John Mettraux - http://lambda.io/jmettraux

-- 
-- 
you received this message because you are subscribed to the "ruote users" group.
to post : send email to [email protected]
to unsubscribe : send email to [email protected]
more options : http://groups.google.com/group/openwferu-users?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"ruote" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to