[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset (KIP-101)

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15957938#comment-15957938 ] Jun Rao commented on KAFKA-1211: The PR for this issue is https://github.com/apache/kafka/pull/2808 >

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15419984#comment-15419984 ] Flavio Junqueira commented on KAFKA-1211: - [~junrao] bq. the leader needs to first wait for the

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15412133#comment-15412133 ] Jun Rao commented on KAFKA-1211: [~fpj], for (a), this is because the leader needs to first wait for the

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15412111#comment-15412111 ] Jun Rao commented on KAFKA-1211: [~fpj], yes, the idea is to always first write the new LGS before writing

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15410693#comment-15410693 ] Flavio Junqueira commented on KAFKA-1211: - [~junrao] I was reading point (a) in your answer again,

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15410682#comment-15410682 ] Flavio Junqueira commented on KAFKA-1211: - [~junrao] I get (2), but (1) might need a bit more

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15406915#comment-15406915 ] Jun Rao commented on KAFKA-1211: [~fpj], very good questions. 1. Yes, the idea is for the follower to

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15406063#comment-15406063 ] Flavio Junqueira commented on KAFKA-1211: - Thanks for the clarification, [~junrao]. There are a

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15404404#comment-15404404 ] Jun Rao commented on KAFKA-1211: [~fpj], for #1 and #2, there are a couple scenarios that this proposal

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15404245#comment-15404245 ] Flavio Junqueira commented on KAFKA-1211: - [~junrao] let me ask a few clarification questions. #

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15402622#comment-15402622 ] Jun Rao commented on KAFKA-1211: The following is a draft proposal. [~fpj], does that look reasonable to

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15318817#comment-15318817 ] Ismael Juma commented on KAFKA-1211: Thanks Jun, I changed the fix version since I don't think we are

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15318799#comment-15318799 ] Jun Rao commented on KAFKA-1211: This is still an issue. It can cause data loss if the leader of a

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15317726#comment-15317726 ] Ismael Juma commented on KAFKA-1211: [~guozhang], is this still an issue? > Hold the produce request

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack > 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15298900#comment-15298900 ] Gwen Shapira commented on KAFKA-1211: - [~junrao] is this still an issue? > Hold the produce request

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14073346#comment-14073346 ] Jun Rao commented on KAFKA-1211: Yes, this is a potential problem. Waiting for HW to be

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14073422#comment-14073422 ] Guozhang Wang commented on KAFKA-1211: -- Jun, I think in your review of KAFKA-1430's

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14073449#comment-14073449 ] Jun Rao commented on KAFKA-1211: The issue is that this problem not only affects ack1, but

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14073478#comment-14073478 ] Guozhang Wang commented on KAFKA-1211: -- What I meant is that the ack=-1 should be

[jira] [Commented] (KAFKA-1211) Hold the produce request with ack 1 in purgatory until replicas' HW has larger than the produce offset

[ https://issues.apache.org/jira/browse/KAFKA-1211?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14074064#comment-14074064 ] Jun Rao commented on KAFKA-1211: Not quite. The case that I described above could happen