Skip to content

Latest commit

 

History

History
130 lines (101 loc) · 5 KB

CHANGELOG.md

File metadata and controls

130 lines (101 loc) · 5 KB
Raw

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

##0.6.7

Changed

  • aiokafka dependency is set to >=0.7.1,<0.8.0

##0.6.6

Bug Fixes

  • Error when logging slow process in aiokafka with python 3.9 153
  • FIX Record instances deserialize properly when returned by agent.ask151

##0.6.5

Bug Fixes

  • Fix leader topic replication 149

##0.6.4

Bug Fixes

  • Fix partition from message for globaltables or tables that use partitioner 129
  • Calling window close callback after popping the value from store 137 ##0.6.3

Bug Fixes

  • Fix for 126 ##0.6.2

Bug Fixes

  • Add app_name to prometheus sensors120 -DatadogMonitor - IndexError: deque index out of range113

##0.6.1

Bug Fixes

  • fix agents with multiple topics116 -Simplify the code by getting rid of deque_pushpopmax in favour of using maxlen= parameter of deque115 -fixes a race condition in writing messages to topics that resulted in a violation of the ordering guarantee (especially changelog topics)112 ##0.6.0

Bug Fixes

-Adding support for aerospike 114

##0.4.7

Bug Fixes

  • Allow multiple workers to share rocksdb data dir98
  • Fix rebalance and recovery issues 83 -92

##0.4.6

Bug Fixes

  • Fix for 85
  • Fix for 88
  • Fix for 91
  • Enabled Tests 79
  • Fix for 84

##0.4.5

Features

  • Peformance improvements in rocksdb by ignore bloom filters within an event context

##0.4.3

Features

  • New Sensor Support to monitor ThreadedProducer 84
  • Support for new rocksdb library faust-streaming-rocksdb 85

##0.4.2

Features

  • New threaded producer for applications using send_soon for sending messages

##0.4.1

Fixed

  • Adding Prometheus Sensor
  • Stability fixes for table recovery when the stream buffer is full 75

##0.4.0

Fixed

  • Prevent stream buffer overflow by lowering the rate of incoming partitions 53 -Recovery thread updating standby partition writes in single writes instead of using writeBatch 51
  • IllegalStateException on seek to offset of a partition that was removed by a rebalance 54

0.3.1

Fixed

-Updating opentracing dependency50 -New-offset is off between 0.2.0 and 0.3.0, resulting in reprocessing last record (or many records) on worker restart 48 -Worker fails to recover table with exactly_once guarantee 47

0.3.0

Fixed

Recovery Crash 44 37

0.2.2

Fixed

  • Consumer offsets not progressing for certain partitions
  • Agent dies silenty when mode cancels pending tasks 678

0.2.1

Fixed

  • Prometheus rebalance typo #26
  • Make SCRAM-SHA-256/512 SASL Mechanisms available #29

0.2.0

Added

  • Hanging of workers on kafka rebalance #21

0.1.1

Fixed

  • web bind to should be 0.0.0.0 instead of localhost #24

0.1.0

Added

  • First release of faust fork
  • Replaced robinhood-aiokafka with aiokafka
  • Implemented transaction support with aiokafka instead of using the the MultiTXNProducer in the robinhood-aiokafka version. Note that this new transaction implementation will create a producer for each kafka group/partition pair