Process max delivery boundaries when expiring vs always putting back on the redelivered queue.
Previously we would always re-queue and do this on getNextMsg() which worked well for push consumers, but with pull based consumers would require a new pull request to be present to process any redelivered and this could report redelivered status incorrectly on max deliver of 1.
Process max delivery boundaries when expiring vs always putting back on the redelivered queue.
Previously we would always re-queue and do this on getNextMsg() which worked well for push consumers, but with pull based consumers would require a new pull request to be present to process any redelivered and this could report redelivered status incorrectly on max deliver of 1.
Signed-off-by: Derek Collison derek@nats.io