Kay Johansen

It seems to happen on a cycle. The offset is stale for a few hours, then the new consumers get the correct offset when they join, for a while, then...

From the client logs. For partition 1, the client received message at 32266 and called commitSync(). But during the rebalance, the consumer is assigned the starting offset 32153. ``` "May...

@Demogorgon314 This seems fixed in 2.9.3.13, does that sound right? Do you know what PR fixes this?

While the client workload was running, I also added a Flink cluster, which may have started the function worker on the broker.

I was able to reproduce the issue after tearing down this cluster and creating a new one with the same name (kj-kop). The broker ran successfully with light load for...

Regarding the LockBusyException, @codelipenghui mentioned that "the broker crashed but the zookeeper lock was not released in time." Do we want to take any action on this? Or have we...

Just an update, I let the workload run about 8 hours today, and my broker restarted 8 times. (which could be 4 OOMs with 2 restarts each time, if it...

The first broker restart is because of OOM, this is captured in issue streamnative/kop#1278. The second broker restart is because of the LockBusyException. I will edit the title of this...

I think the maintainer of zlib has made a fix in the code, but has not made a release containing the fix https://github.com/madler/zlib/issues/686

Here the client receives the same message `1:157137` three times [Client logs](https://kibana.sncloud-stg.dev/app/discover#/view/dd5f7250-4b20-11ed-8cc3-0f6424602b19?_g=(filters:!(),refreshInterval:(pause:!t,value:0),time:(from:'2022-10-17T06:33:00.524Z',to:'2022-10-17T10:34:21.018Z'))&_a=(columns:!(kubernetes.pod_name,message),filters:!(),index:dec82910-8a4c-11ec-85bd-2f0e9c442e06,interval:auto,query:(language:kuery,query:'kubernetes.pod_namespace:%22sn-cv%22%20AND%20kubernetes.pod_name:%20kop-produce-consume-test-2-9-kop2*%20and%20message:157137'),sort:!(!('@timestamp',desc)))) ``` Received message 1:157709, 1666002785014 with latency of 505 ms Resetting offset for partition public/default/kop-topic-1 to position FetchPosition{offset=156746...