Perhaps I am misunderstanding 'auto complete' but is it not 'safer' to only release the msg from the Q after it has been successfully processed by the logic App (i.e. use 'peak lock')?
Thanks for the question. There are two approaches: the "release" controlled by the consumer (effectively - connector to the queue/topic) and "release" controlled by your own code running in the consumer. Both work equally well if your custom code comply the rules. Btw, "release" means - return the massage to the queue to be processed again. "Autom-complete" will always return the message into the queue/topic in case an unhandled error occurred. Using "peek-lock" you make your code more flexible about deciding when a message has to be released or completed. You may release and return the message into the queue even though there was no error in the consumer handler. Having said this, it depends, you decide. Auto-complete is simpler to use.
Great session Sr. Maxim thanks a lot!
My pleasure. You may find something else helping on my channel and it is alive, there are more to come.
Perhaps I am misunderstanding 'auto complete' but is it not 'safer' to only release the msg from the Q after it has been successfully processed by the logic App (i.e. use 'peak lock')?
Thanks for the question. There are two approaches: the "release" controlled by the consumer (effectively - connector to the queue/topic) and "release" controlled by your own code running in the consumer. Both work equally well if your custom code comply the rules. Btw, "release" means - return the massage to the queue to be processed again. "Autom-complete" will always return the message into the queue/topic in case an unhandled error occurred. Using "peek-lock" you make your code more flexible about deciding when a message has to be released or completed. You may release and return the message into the queue even though there was no error in the consumer handler.
Having said this, it depends, you decide. Auto-complete is simpler to use.