klevy-toast commented on issue #19134: URL: https://github.com/apache/pulsar/issues/19134#issuecomment-1416567954
> If you are observing this behavior, that sounds like a bug. It looks like there might be a race in the logic, but the client does attempt to handle this case here: > > https://github.com/apache/pulsar/blob/39dd1cda2d01a6d472b7a39115a774958a837be1/pulsar-client/src/main/java/org/apache/pulsar/client/impl/ConsumerImpl.java#L746-L752 I don't _think_ this behavior is necessarily a bug. What I mean by that, is if my application takes >30s to process a message, then the ackTimeout setting kicks in and tells pulsar to redeliver it. Then the pulsar arrives to my application again (likely on a different host), and again takes >30s to redeliver. We aren't getting an opportunity to nack the message, but instead just causing pulsar to redeliver our message over & over. We observed this a lot on the 2.7.2 client, so maybe it has improved since then -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
