Hello all,
I am developing a manual retry pattern, involving a collection of
shapes including some delay shapes for controlling retries, some scopes
for error handler and some variables to control these. Finally i want
to implement a Suspend Shape when all retries are exhausted.

The orchestration is becoming suspended, as i can see in the bts2k6
admin, it says it is suspended but resumable, and i resubmit the
instance from there and i am expecting to reprocess the service
orchestration from the point of failure which is the scope i've
declared.

When i resume the instance, it is supposedly get on work again, but
when i check the hat the instance is still running, and in the
orchestration debugger is reached only the suspend shape but ignoring
the resubmitting part.

The instance is kept active all the way, it never completes its work
eventhough i resumed the instance and make sure no errors raised, so i
expect after resubmitting it appears as completed. I have to terminate
manually the instance. I must mention that it is erased from the
suspended instances, but it is in the active and therefore i cannot
undeploy or anything.

How does the suspend shape works ? If any, where will be resumed, from
the point if failure or from the beginning of the orchestration/scope ?

Thank you in advance !!

[ Post a follow-up to this message ]

You need to implement resumability "manually" too. Place the processing logc
in a loop where the loop continues till say, a bool variable "isComplete" is
set to True. If your retry succeeds, set this flag to true but if retries
fail, let it remain false. When you resume the orchestration, it will attemp
t
to go to the point at which it suspended - so it goes back to the loop, wher
e
the condition "isComplete = False" is still true, and that's how processing
starts again. If you're using retry counters, you must reset the counter to
zero before the suspend shape.

"rene.rugerio@gmail.com" wrote:

> Hello all,
> I am developing a manual retry pattern, involving a collection of
> shapes including some delay shapes for controlling retries, some scopes
> for error handler and some variables to control these. Finally i want
> to implement a Suspend Shape when all retries are exhausted.
>
> The orchestration is becoming suspended, as i can see in the bts2k6
> admin, it says it is suspended but resumable, and i resubmit the
> instance from there and i am expecting to reprocess the service
> orchestration from the point of failure which is the scope i've
> declared.
>
> When i resume the instance, it is supposedly get on work again, but
> when i check the hat the instance is still running, and in the
> orchestration debugger is reached only the suspend shape but ignoring
> the resubmitting part.
>
> The instance is kept active all the way, it never completes its work
> eventhough i resumed the instance and make sure no errors raised, so i
> expect after resubmitting it appears as completed. I have to terminate
> manually the instance. I must mention that it is erased from the
> suspended instances, but it is in the active and therefore i cannot
> undeploy or anything.
>
> How does the suspend shape works ? If any, where will be resumed, from
> the point if failure or from the beginning of the orchestration/scope ?
>
> Thank you in advance !!
>
>

[ Post a follow-up to this message ]