1. References
  2. StreamNative Pulsar Changelogs
  3. v3.0

StreamNative Weekly Release Notes v3.0.0.4

StreamNative Weekly Release Notes v3.0.0.4



  • https://github.com/streamnative/pulsar/releases/tag/v3.0.0.4



General Changes

Apache Pulsar

(#20597))) Revert "[fix][broker] Fix NPE when reset Replicator's cursor by position.

(#20710) [fix][broker] Revert "Skip loading broker interceptor when disableBrokerInterceptors is true #20422"

(#20666) [improve][broker] Make ExtensibleLoadManagerImpl's broker filter pure async

(#20609) [fix][test] Fix flaky test ExtensibleLoadManagerTest.testIsolationPolicy

(#20138) [improve][test] Add integration test for ExtensibleLoadManager

(#20677) [fix][broker] Added the skipped message handler for ServiceUnitStateChannel

(#20595) [fix][client] Make the whole grabCnx() progress atomic

(#20652) [improve][broker] Make get list from bundle Admin API async

(#20605) [fix][sql] Remove useless configuration for Pulsar SQL

(#20698) [fix][io] Close the kafka source connector got stuck

(#20569) [fix][txn] Use PulsarResource check for topic existence instead of brokerservice.getTopic()

(#20365) [fix][offload] Filesystem offloader class not found hadoop-hdfs-client

(#19775) [fix] [Perf] PerformanceProducer do not produce expected number of messages.

(#20457) [fix][broker] Fix return the earliest position when query position by timestamp.

(#20613) [fix][broker] Topic policy can not be work well if replay policy message has any exception.

(#20597) [fix][broker] Fix NPE when reset Replicator's cursor by position.

(#20642) [fix][meta] Bookie Info lost by notification race condition.

(#20641) [fix][meta] Adding the missed bookie id in the registration manager.

(#20704) [improve][broker] Upgrade bookkeeper to 4.16.2

(#20607) [fix][broker]fix the publish latency spike issue with large number of producers

(#20689) [fix][fn] Exit JVM when main thread throws exception

(#20669) [fix][client]Fix deadlock issue of consumer while using multiple IO threads

(#20647) [fix][broker] Fix the publish latency spike from the contention of MessageDeduplication

(#20690) [fix] [admin] set ns level backlog quota does not take effect if retention exists

(#20671) [improve][test] Disable disk usage threshold & geoip download and enable logging for Elastic Testcontainers

(#20149) [improve][bk] Add integration test with bookie http server enabled

(#16832) [improve][broker] Support cgroup v2 by using jdk.internal.platform.Metrics in Pulsar Loadbalancer


(#939) [ci] Ignore the jms1_1 test first


(#1923) [transaction] Implement producer state manager recovery

(#1914) Fix flaky OffsetTopicWriteTimeoutTest

(#1907) [transactions] Better handling of network exceptions while sending TX markers

(#1906) [transaction] Producer state manager snapshot recovery - Part-1: Add snapshot I/O buffer

(#1905) Reduce the offset commit timeout to 5 seconds and make it configurable

(#1903) Fix flaky CompactedPartitionedTopicTest.testClose

(#1899) [improvements] Use the TopicEventListener API

(#1893) [improvement] Remove expensive useless String.format() in canConsumeAsync

StreamNative Pulsar Plugins

[fix][test] Reinstall the component ca-certificates-java in the DockerFile of detector test

Function Mesh Worker Service

Support manual semantics