You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Oct 17, 2018. It is now read-only.
When there is a very slow consumer, the producer could fill up the tcp write buffer and time out on tcp writes, which will cause the producer to close the connection and recreate a new one.
But there is no way for consumer to know that the existing connection is closed, which means it will continue to read msg from the connection and work on the msg, although all of the work will be wasted as they won't be acked back to the producer, so the messages will be tried again.
It would be good to somehow detect the close of the connection and stop processing messages from it.
The text was updated successfully, but these errors were encountered:
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
When there is a very slow consumer, the producer could fill up the tcp write buffer and time out on tcp writes, which will cause the producer to close the connection and recreate a new one.
But there is no way for consumer to know that the existing connection is closed, which means it will continue to read msg from the connection and work on the msg, although all of the work will be wasted as they won't be acked back to the producer, so the messages will be tried again.
It would be good to somehow detect the close of the connection and stop processing messages from it.
The text was updated successfully, but these errors were encountered: