Refer to the exhibit. This Mule application is deployed to multiple CloudHub workers with persistent queues enabled.
The retrieveFile flow's event source reads a CSV file from a remote SFTP server and then publishes each record in the CSV file to a VM queue.
The processCustomerRecords flow's VM Listener receives messages from the same VM queue and then processes each message separately.
How are messages routed to the CloudHub workers as messages are received by the VM Listener?
Each message is routed to one of the available CloudHub workers in a non-deterministic non-round-robin fashion, thereby approximately balancing messages among the CloudHub workers. When VM queues are used with multiple workers in a Mule application deployed on CloudHub, messages are not distributed in a deterministic or round-robin manner, but rather in a way that aims to balance the load approximately across the available workers. This non-deterministic distribution helps in balancing the processing load under varying conditions.
C. Each message is routed to ONE of the available CloudHub workers in a NON-DETERMINISTIC non-round-robin fashion, thereby APPROXIMATELY BALANCING messages among the CloudHub workers
It should be C. I was leaning towards D but D is only applicable if we are dealing with HTTP requests. The question is about Persistent VM queue consumer. Because its Persistent VM which is going to distribute message across workers in non-deterministic way hence causing messages to be processed in non-deterministic way.
Agree with C. ref - https://help.mulesoft.com/s/question/0D52T00004mXXXSSA4/vm-deployed-into-cloudhub-with-multiple-workers