[ https://issues.apache.org/jira/browse/FLINK-20114?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17896853#comment-17896853 ]
Jiangjie Qin commented on FLINK-20114: -------------------------------------- {quote}why the _enable.auto.commit_ behavior turned off by default (The Kafka officially says defaults to true if there group.id provided) which may cause user use must define expilicity true to the kafka source DDL. (naturely peolple won't set it){quote} The idea is to associate the offset commit with the Flink checkpoint to avoid the rewind in committed offset. So the Flink job will only commit an offset after a checkpoint has succeeded. Otherwise, what could happen is after a job failover, the committed offset may go back because the job resets to the last successful checkpoint. By default checkpoint is enabled in Flink. So by default the offsets will be committed to Kafka. When checkpoint is disabled, to make sure of the same semantic of when checkpoint is enabled, by default no offset is committed unless user explicitly specifies to do so. But I can understand the argument that if checkpoint is disabled, default Kafka behavior should be honored. So both behavior has their own point. But from backwards compatibility perspective, I think keep the behavior the same as {{LiKafkaConsumer}} makes sense. {quote}It seems that this ticket takes effects earlier than 1.13.5, what causes the commiting to consumer group behaivor diffierence between Flink 1.13.5 and Flink 1.17.2. (no checkpoint enabled)?{quote} I am not sure about this. > Fix a few KafkaSource-related bugs > ---------------------------------- > > Key: FLINK-20114 > URL: https://issues.apache.org/jira/browse/FLINK-20114 > Project: Flink > Issue Type: Bug > Components: Connectors / Kafka > Affects Versions: 1.12.0 > Reporter: Robert Metzger > Priority: Critical > Labels: pull-request-available, release-testing > Fix For: 1.13.0, 1.12.4 > > Attachments: Screenshot 2020-11-24 at 15.14.35.png, first-run.tgz, > second-run.tgz > > > Feature introduced in https://issues.apache.org/jira/browse/FLINK-18323 > -------- > [General Information about the Flink 1.12 release > testing|https://cwiki.apache.org/confluence/display/FLINK/1.12+Release+-+Community+Testing] > When testing a feature, consider the following aspects: > - Is the documentation easy to understand > - Are the error messages, log messages, APIs etc. easy to understand > - Is the feature working as expected under normal conditions > - Is the feature working / failing as expected with invalid input, induced > errors etc. > If you find a problem during testing, please file a ticket > (Priority=Critical; Fix Version = 1.12.0), and link it in this testing ticket. > During the testing, and once you are finished, please write a short summary > of all things you have tested. -- This message was sent by Atlassian Jira (v8.20.10#820010)