There is important data in the info messages without which its difficult after the event to understand what MTS a computation was processing at runtime. Exposing the info messages allows consumers to do what they want with the message e.g. expose rollup and MTS counts in their own observability by parsing from the info messages.
I know this breaks the model used by say the limit size where the message is consumed and stored in a client property but its necessary for things like MTS counts processed by the data since it needs to be consumed on a thread watching the channel rather than a process interrogating client properties that store the info.
There is important data in the info messages without which its difficult after the event to understand what MTS a computation was processing at runtime. Exposing the info messages allows consumers to do what they want with the message e.g. expose rollup and MTS counts in their own observability by parsing from the info messages. I know this breaks the model used by say the limit size where the message is consumed and stored in a client property but its necessary for things like MTS counts processed by the data since it needs to be consumed on a thread watching the channel rather than a process interrogating client properties that store the info.