Azure Service Bus queue partitioned Queue AutoRenewTimeout

Question

I want to understand if AutoRenewTimeout in case of a partitioned queue is different from not partitioned one. I have two reasons to ask this here:

1) They seem to have different explanation in documentation:

Partitioned Queue - link

Gets or sets the time needed before the session renew its state.

Non Partitioned Queue - link

Gets or sets the maximum duration within which the lock will be renewed automatically. This value should be greater than the longest message lock duration; for example, the LockDuration Property.

2) After enabling former on my code, with lock duration = 4 minutes and AutoRenewTimeout = 12 minutes, I'm seeing much lower productivity.

Do they have different meaning in both contexts?

Answer 1

The first link is for Message Session feature. With message sessions, only one instance can process messages from the same session. Which could explain why you've seen lower performance. The second option doesn't leverage message session and therefore messages are consumed by all the competing consumers, rather than just a single one.