Re: [DISCUSS] KIP-1043: Administration of groups

Hi Apoorv, Thanks for your comments. AM1: I chose to leave the majority of the administration for the different types of groups in their own tools. The differences between the group types are significant and I think that one uber tool that subsumes kafka-consumer-groups.sh, kafka-share-groups.sh a

[jira] [Resolved] (KAFKA-17121) junit-platform.properties files in published artifacts pollute the test classpath of consumers

[ https://issues.apache.org/jira/browse/KAFKA-17121?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Chia-Ping Tsai resolved KAFKA-17121. Fix Version/s: 3.9.0 Resolution: Fixed > junit-platform.properties files in publish

[jira] [Resolved] (KAFKA-17132) Revisit testMissingOffsetNoResetPolicy for AsyncConsumer

[ https://issues.apache.org/jira/browse/KAFKA-17132?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Chia-Ping Tsai resolved KAFKA-17132. Fix Version/s: 3.9.0 Resolution: Fixed > Revisit testMissingOffsetNoResetPolicy for

Re: [DISCUSS] KIP-1071: Streams Rebalance Protocol

Hi Andrew, thanks for getting the discussion going! Here are my responses. AS1: Good point, done. AS2: We were planning to add more administrative tools to the interface in a follow-up KIP, to not make this KIP too large. If people think that it would help to understand the overall picture if we

[jira] [Resolved] (KAFKA-12877) Fix KRPC files with missing flexibleVersions annotation

[ https://issues.apache.org/jira/browse/KAFKA-12877?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Chia-Ping Tsai resolved KAFKA-12877. Fix Version/s: 3.0.0 Resolution: Fixed > Fix KRPC files with missing flexibleVersio

Re: [DISCUSS] KIP-1068: KIP-1068: New JMX Metrics for AsyncKafkaConsumer

Hi Brendan, Thanks for the KIP. The metrics are always helpful. AM1: Is `application-event-queue-age-avg` enough or do we require ` application-event-queue-age-max` as well to differentiate with outliers? AM2: The kafka producer defines metric `record-queue-time-avg` which captures the time spent

Re: [DISCUSS] KIP-1071: Streams Rebalance Protocol

Hi Lucas, Thanks for your response. All makes sense for me, with just a couple of follow-up comments. AS8: So, really the broker config is the name of the default assignor used unless it’s overridden by a group config. I have one suggestion, which you can of course ignore, that you use `group.stre

Re: [DISCUSS] KIP-1071: Streams Rebalance Protocol

Hi Lucas, I see that I hit send too quickly. One more comment: AS2: I think stating that there will be a `kafka-streams-group.sh` in a future KIP is fine to keep this KIP focused. Personally, I would probably put all of the gory details in this KIP, but then it’s not my KIP. A future pointer is fi

Re: [DISCUSS] KIP-1071: Streams Rebalance Protocol

Hi Andrew, AS2: I added a note for now. If others feel strongly about it, we can still add more administrative tools to the KIP - it should not change the overall story significantly. AS8: "streams.group.assignor.name" sounds good to me to distinguish the config from class names. Not sure if I li

[jira] [Created] (KAFKA-17169) Add an EndpointsTest suite that test all of the public methods

José Armando García Sancio created KAFKA-17169: -- Summary: Add an EndpointsTest suite that test all of the public methods Key: KAFKA-17169 URL: https://issues.apache.org/jira/browse/KAFKA-17169

Re: [DISCUSS] KIP-1068: KIP-1068: New JMX Metrics for AsyncKafkaConsumer

Hi Apoorv, Thank you for your comments, I will address each. AM1. I can see the usefulness in also having an 'application-event-queue-age-max' to get an idea of outliers and how they may be affecting the average metric. I will add that. AM2. I agree with you there, I think 'time' is a better desc

Re: [DISCUSS] KIP-1059: Enable the Producer flush() method to clear the latest send() error

For catching client side errors this would work IMHO. I am ok with this. We throw before we add the record to the batch. Very clean semantics which should also address the concern of "non-atomic tx"... The exception clearly indicates that the record was not added to the TX, and users can react

Re: [DISCUSS] KIP-1071: Streams Rebalance Protocol

Hi Lucas/Bruno, thanks for the great KIP! First comments: LM1. Related to where the KIP says: *“Group ID, member ID, member epoch are sent with each heartbeat request. Any other information that has not changed since the last heartbeat can be omitted.”. *I expect all the other info also needs to

[jira] [Created] (KAFKA-17170) Add test to ensure new consumer acks reconciled assignment even if first HB with ack lost

Lianet Magrans created KAFKA-17170: -- Summary: Add test to ensure new consumer acks reconciled assignment even if first HB with ack lost Key: KAFKA-17170 URL: https://issues.apache.org/jira/browse/KAFKA-17170

Re: [DISCUSS] KIP-1059: Enable the Producer flush() method to clear the latest send() error

Hi Artem and Matthias, > On the other hand, the effort to prove that > keeping all records in memory won't break some scenarios (and generally > breaking one is enough to cause a lot of pain) seems to be significantly > higher than to prove that setting some flag in some API has pretty much 0 > ch