Alexey Biryukov
Alexey Biryukov
**Triage bottomline**: failure due to an issue in the test. Consumer groups described in the source and in the target cluster are equal with the exception of the order of...
ASAN has detected *heap-use-after-free* during RP shutdown, when many services including `raft::group_manager` have been stopped and destroyed, but reactor still runs. Adopted stack trace: ``` std::exception_ptr::operator bool() ss::basic_semaphore::signal(unsigned long) at...
**Triage bottomline**: extremely likely an RP bug, medium or low severity because it apparenlty happens during shutdowm only and because it is rare.
The failure is resulted by a really rare combination of factors specific to the test. One of the nodes undergoing maintenance restart and unavailable while another node serves a fatch...
May be related: [KAFKA-10313](https://issues.apache.org/jira/browse/KAFKA-10313)
[5788](https://github.com/redpanda-data/redpanda/pull/5788) may help with more context
The issue manifests itself in the MirrorMaker test. The list of partitions inside RP is indeed stable, however when partitions are copied over to Kafka with MirrorMaker, the order of...
The test runs 2 consumers in the same group. The failing criteria is supposed to verify that each consumer consumes at least 50 messages out of 5000 from the topic....
https://github.com/redpanda-data/redpanda/issues/5275 is a duplicate of this one
The test is run on 3 nodes. The sequence is: create topic - produce and consume in a group - delete topic - restart all nodes - * - create...