Ron Ten-Hove wrote:
> Satish Thatte wrote:
>
>> What did you have in mind for the handling of concurrency faults?
>> Retry? Failure? Some custom code?
>>
> This is really a specific case of the more general problem of fault
> handling / recovery in BPEL. The language should have decent support
> for this. In this case, I might follow a typical pattern: compensate
> for the activities completed in the failed scope, then retry it.
> Nothing fancy like custom code, as much as I admire Edwin's <exec>
> tags :-).
Hum. Upon reflection, I think the error handling might be more difficult
than I originally thought. The fact that a serializable scope must be a
leaf scope makes structuring error handling within the scope in question
more difficult. This brings up the question: what would be the cost of
allowing nested scopes within a serializable one? Effectively it would
function as one collective serializable scope, but would make
structuring fault and compensation handling within the serializable
scope more flexible.
Also, upon further relection, what did you mean by "custom code"? Did
you mean a user-defined fault handler for concurrency faults?
-Ron
