[jira] [Created] (KAFKA-7504) Broker performance degradation caused by call of sendfile reading disk in network thread

Yuto Kawamura created KAFKA-7504: Summary: Broker performance degradation caused by call of sendfile reading disk in network thread Key: KAFKA-7504 URL: https://issues.apache.org/jira/browse/KAFKA-7504

[jira] [Created] (KAFKA-3775) Throttle maximum number of tasks assigned to a single KafkaStreams

Yuto Kawamura created KAFKA-3775: Summary: Throttle maximum number of tasks assigned to a single KafkaStreams Key: KAFKA-3775 URL: https://issues.apache.org/jira/browse/KAFKA-3775 Project: Kafka

[jira] [Work started] (KAFKA-3775) Throttle maximum number of tasks assigned to a single KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3775?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Work on KAFKA-3775 started by Yuto Kawamura. > Throttle maximum number of tasks assigned to a single KafkaStreams > --

[jira] [Updated] (KAFKA-3775) Throttle maximum number of tasks assigned to a single KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3775?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3775: - Status: Patch Available (was: In Progress) > Throttle maximum number of tasks assigned to a singl

[jira] [Commented] (KAFKA-3775) Throttle maximum number of tasks assigned to a single KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3775?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15311838#comment-15311838 ] Yuto Kawamura commented on KAFKA-3775: -- [~guozhang] What do you think? > Throttle ma

[jira] [Commented] (KAFKA-3775) Throttle maximum number of tasks assigned to a single KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3775?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15313577#comment-15313577 ] Yuto Kawamura commented on KAFKA-3775: -- Thanks for feedback [~mjsax] . > 1) a KStrea

[jira] [Commented] (KAFKA-3775) Throttle maximum number of tasks assigned to a single KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3775?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15313828#comment-15313828 ] Yuto Kawamura commented on KAFKA-3775: -- Thanks for feedback [~BigAndy] . > With the

[jira] [Comment Edited] (KAFKA-3775) Throttle maximum number of tasks assigned to a single KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3775?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15313828#comment-15313828 ] Yuto Kawamura edited comment on KAFKA-3775 at 6/3/16 8:17 AM: --

[jira] [Commented] (KAFKA-3775) Throttle maximum number of tasks assigned to a single KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3775?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15324187#comment-15324187 ] Yuto Kawamura commented on KAFKA-3775: -- Sorry for leaving this discussion for a while

[jira] [Commented] (KAFKA-3775) Throttle maximum number of tasks assigned to a single KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3775?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15326484#comment-15326484 ] Yuto Kawamura commented on KAFKA-3775: -- [~guozhang] Yes, both of your short and long

[jira] [Updated] (KAFKA-3775) Throttle maximum number of tasks assigned to a single KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3775?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3775: - Resolution: Won't Fix Status: Resolved (was: Patch Available) > Throttle maximum number o

[jira] [Closed] (KAFKA-3775) Throttle maximum number of tasks assigned to a single KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3775?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura closed KAFKA-3775. > Throttle maximum number of tasks assigned to a single KafkaStreams > -

[jira] [Created] (KAFKA-3947) kafka-reassign-partitions.sh should support dumping current assignment

Yuto Kawamura created KAFKA-3947: Summary: kafka-reassign-partitions.sh should support dumping current assignment Key: KAFKA-3947 URL: https://issues.apache.org/jira/browse/KAFKA-3947 Project: Kafka

[jira] [Work started] (KAFKA-3947) kafka-reassign-partitions.sh should support dumping current assignment

[ https://issues.apache.org/jira/browse/KAFKA-3947?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Work on KAFKA-3947 started by Yuto Kawamura. > kafka-reassign-partitions.sh should support dumping current assignment > --

[jira] [Updated] (KAFKA-3947) kafka-reassign-partitions.sh should support dumping current assignment

[ https://issues.apache.org/jira/browse/KAFKA-3947?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3947: - Status: Patch Available (was: In Progress) > kafka-reassign-partitions.sh should support dumping

[jira] [Updated] (KAFKA-3947) kafka-reassign-partitions.sh should support dumping current assignment

[ https://issues.apache.org/jira/browse/KAFKA-3947?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3947: - Description: When I building my own tool to perform reassignment of partitions, I realized that t

[jira] [Created] (KAFKA-4024) First metadata update always take retry.backoff.ms milliseconds to complete

Yuto Kawamura created KAFKA-4024: Summary: First metadata update always take retry.backoff.ms milliseconds to complete Key: KAFKA-4024 URL: https://issues.apache.org/jira/browse/KAFKA-4024 Project: Ka

[jira] [Commented] (KAFKA-4024) First metadata update always take retry.backoff.ms milliseconds to complete

[ https://issues.apache.org/jira/browse/KAFKA-4024?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15410560#comment-15410560 ] Yuto Kawamura commented on KAFKA-4024: -- I guess this problem is fairly easy to fix, j

[jira] [Updated] (KAFKA-4024) First metadata update always take retry.backoff.ms milliseconds to complete

[ https://issues.apache.org/jira/browse/KAFKA-4024?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-4024: - Status: Patch Available (was: Open) > First metadata update always take retry.backoff.ms millisec

[jira] [Commented] (KAFKA-4024) First metadata update always take retry.backoff.ms milliseconds to complete

[ https://issues.apache.org/jira/browse/KAFKA-4024?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15410571#comment-15410571 ] Yuto Kawamura commented on KAFKA-4024: -- Submitted a patch. It actually leads a side e

[jira] [Commented] (KAFKA-4024) First metadata update always take retry.backoff.ms milliseconds to complete

[ https://issues.apache.org/jira/browse/KAFKA-4024?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15410572#comment-15410572 ] Yuto Kawamura commented on KAFKA-4024: -- [~ijuma] or [~hachikuji] PTAL. > First metad

[jira] [Updated] (KAFKA-4710) Logging trait breaking logging location information retrival by log4j

[ https://issues.apache.org/jira/browse/KAFKA-4710?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-4710: - Status: Open (was: Patch Available) > Logging trait breaking logging location information retriva

[jira] [Created] (KAFKA-4614) Long GC pause harming broker performance which is caused by mmap objects created for OffsetIndex

Yuto Kawamura created KAFKA-4614: Summary: Long GC pause harming broker performance which is caused by mmap objects created for OffsetIndex Key: KAFKA-4614 URL: https://issues.apache.org/jira/browse/KAFKA-4614

[jira] [Comment Edited] (KAFKA-4614) Long GC pause harming broker performance which is caused by mmap objects created for OffsetIndex

[ https://issues.apache.org/jira/browse/KAFKA-4614?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15817627#comment-15817627 ] Yuto Kawamura edited comment on KAFKA-4614 at 1/11/17 8:19 AM: -

[jira] [Commented] (KAFKA-4614) Long GC pause harming broker performance which is caused by mmap objects created for OffsetIndex

[ https://issues.apache.org/jira/browse/KAFKA-4614?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15817627#comment-15817627 ] Yuto Kawamura commented on KAFKA-4614: -- I've tried to fix this and now I'm testing pa

[jira] [Commented] (KAFKA-4614) Long GC pause harming broker performance which is caused by mmap objects created for OffsetIndex

[ https://issues.apache.org/jira/browse/KAFKA-4614?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15820948#comment-15820948 ] Yuto Kawamura commented on KAFKA-4614: -- I applied the patch to one of our production

[jira] [Updated] (KAFKA-4614) Long GC pause harming broker performance which is caused by mmap objects created for OffsetIndex

[ https://issues.apache.org/jira/browse/KAFKA-4614?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-4614: - Status: Patch Available (was: Open) > Long GC pause harming broker performance which is caused by

[jira] [Updated] (KAFKA-4614) Long GC pause harming broker performance which is caused by mmap objects created for OffsetIndex

[ https://issues.apache.org/jira/browse/KAFKA-4614?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-4614: - Attachment: kafka-produce-99th.png Just as a sequel, this is what happened to the 99th percentile

[jira] [Comment Edited] (KAFKA-4614) Long GC pause harming broker performance which is caused by mmap objects created for OffsetIndex

[ https://issues.apache.org/jira/browse/KAFKA-4614?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15851093#comment-15851093 ] Yuto Kawamura edited comment on KAFKA-4614 at 2/3/17 5:37 AM: --

[jira] [Comment Edited] (KAFKA-4614) Long GC pause harming broker performance which is caused by mmap objects created for OffsetIndex

[ https://issues.apache.org/jira/browse/KAFKA-4614?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15851093#comment-15851093 ] Yuto Kawamura edited comment on KAFKA-4614 at 2/3/17 5:41 AM: --

[jira] [Comment Edited] (KAFKA-4614) Long GC pause harming broker performance which is caused by mmap objects created for OffsetIndex

[ https://issues.apache.org/jira/browse/KAFKA-4614?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15851093#comment-15851093 ] Yuto Kawamura edited comment on KAFKA-4614 at 2/3/17 5:42 AM: --

[jira] [Created] (KAFKA-3471) min.insync.replicas isn't respected when there's a delaying follower who still in ISR

Yuto Kawamura created KAFKA-3471: Summary: min.insync.replicas isn't respected when there's a delaying follower who still in ISR Key: KAFKA-3471 URL: https://issues.apache.org/jira/browse/KAFKA-3471 P

[jira] [Updated] (KAFKA-3471) min.insync.replicas isn't respected when there's a delaying follower who still in ISR

[ https://issues.apache.org/jira/browse/KAFKA-3471?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3471: - Description: h2. tl;dr; Partition.checkEnoughReplicasReachOffset should see the number of follower

[jira] [Updated] (KAFKA-3471) min.insync.replicas isn't respected when there's a delaying follower who still in ISR

[ https://issues.apache.org/jira/browse/KAFKA-3471?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3471: - Description: tl;dr; Partition.checkEnoughReplicasReachOffset should see the number of followers

[jira] [Updated] (KAFKA-3471) min.insync.replicas isn't respected when there's a delaying follower who still in ISR

[ https://issues.apache.org/jira/browse/KAFKA-3471?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3471: - Status: Patch Available (was: Open) > min.insync.replicas isn't respected when there's a delaying

[jira] [Commented] (KAFKA-3471) min.insync.replicas isn't respected when there's a delaying follower who still in ISR

[ https://issues.apache.org/jira/browse/KAFKA-3471?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15215406#comment-15215406 ] Yuto Kawamura commented on KAFKA-3471: -- [~becket_qin] Ah.. I wasn't aware of the lead

[jira] [Commented] (KAFKA-3471) min.insync.replicas isn't respected when there's a delaying follower who still in ISR

[ https://issues.apache.org/jira/browse/KAFKA-3471?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15219267#comment-15219267 ] Yuto Kawamura commented on KAFKA-3471: -- Thank you very much for your advice. It was v

[jira] [Closed] (KAFKA-3471) min.insync.replicas isn't respected when there's a delaying follower who still in ISR

[ https://issues.apache.org/jira/browse/KAFKA-3471?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura closed KAFKA-3471. > min.insync.replicas isn't respected when there's a delaying follower who > still in ISR > ---

[jira] [Updated] (KAFKA-3471) min.insync.replicas isn't respected when there's a delaying follower who still in ISR

[ https://issues.apache.org/jira/browse/KAFKA-3471?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3471: - Status: Open (was: Patch Available) > min.insync.replicas isn't respected when there's a delaying

[jira] [Resolved] (KAFKA-3471) min.insync.replicas isn't respected when there's a delaying follower who still in ISR

[ https://issues.apache.org/jira/browse/KAFKA-3471?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura resolved KAFKA-3471. -- Resolution: Not A Bug > min.insync.replicas isn't respected when there's a delaying follower who

[jira] [Updated] (KAFKA-3497) Streams ProcessorContext should support forward() based on child name

[ https://issues.apache.org/jira/browse/KAFKA-3497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3497: - Description: Currently {{ProcessorContext}} only supports {{forward(K, V)}} which forwards KV to

[jira] [Created] (KAFKA-3497) Streams ProcessorContext should support forward() based on child name

Yuto Kawamura created KAFKA-3497: Summary: Streams ProcessorContext should support forward() based on child name Key: KAFKA-3497 URL: https://issues.apache.org/jira/browse/KAFKA-3497 Project: Kafka

[jira] [Created] (KAFKA-3616) Make kafka producers/consumers injectable for KafkaStreams

Yuto Kawamura created KAFKA-3616: Summary: Make kafka producers/consumers injectable for KafkaStreams Key: KAFKA-3616 URL: https://issues.apache.org/jira/browse/KAFKA-3616 Project: Kafka Issu

[jira] [Updated] (KAFKA-3616) Make kafka producers/consumers injectable for KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3616?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3616: - Status: Patch Available (was: Open) > Make kafka producers/consumers injectable for KafkaStreams

[jira] [Commented] (KAFKA-3616) Make kafka producers/consumers injectable for KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3616?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15255615#comment-15255615 ] Yuto Kawamura commented on KAFKA-3616: -- https://github.com/apache/kafka/pull/1264 [~g

[jira] [Commented] (KAFKA-3616) Make kafka producers/consumers injectable for KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3616?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15256187#comment-15256187 ] Yuto Kawamura commented on KAFKA-3616: -- > About "reliable message delivery", out of c

[jira] [Issue Comment Deleted] (KAFKA-3616) Make kafka producers/consumers injectable for KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3616?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3616: - Comment: was deleted (was: > About "reliable message delivery", out of curiosity could you elabora

[jira] [Commented] (KAFKA-3616) Make kafka producers/consumers injectable for KafkaStreams

[ https://issues.apache.org/jira/browse/KAFKA-3616?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15256190#comment-15256190 ] Yuto Kawamura commented on KAFKA-3616: -- > About "reliable message delivery", out of c

[jira] [Created] (KAFKA-3642) Fix NPE from ProcessorStateManager when the changelog topic not exists

Yuto Kawamura created KAFKA-3642: Summary: Fix NPE from ProcessorStateManager when the changelog topic not exists Key: KAFKA-3642 URL: https://issues.apache.org/jira/browse/KAFKA-3642 Project: Kafka

[jira] [Updated] (KAFKA-3642) Fix NPE from ProcessorStateManager when the changelog topic not exists

[ https://issues.apache.org/jira/browse/KAFKA-3642?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-3642: - Status: Patch Available (was: Open) > Fix NPE from ProcessorStateManager when the changelog topic

[jira] [Created] (KAFKA-4116) Specifying 0.0.0.0 in "listeners" doesn't work

Yuto Kawamura created KAFKA-4116: Summary: Specifying 0.0.0.0 in "listeners" doesn't work Key: KAFKA-4116 URL: https://issues.apache.org/jira/browse/KAFKA-4116 Project: Kafka Issue Type: Bug

[jira] [Updated] (KAFKA-4116) Specifying 0.0.0.0 in "listeners" doesn't work

[ https://issues.apache.org/jira/browse/KAFKA-4116?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-4116: - Status: Patch Available (was: Open) > Specifying 0.0.0.0 in "listeners" doesn't work > --

[jira] [Commented] (KAFKA-4116) Specifying 0.0.0.0 in "listeners" doesn't work

[ https://issues.apache.org/jira/browse/KAFKA-4116?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15457774#comment-15457774 ] Yuto Kawamura commented on KAFKA-4116: -- [~gwenshap] PTAL. > Specifying 0.0.0.0 in "l

[jira] [Updated] (KAFKA-4024) First metadata update always take retry.backoff.ms milliseconds to complete

[ https://issues.apache.org/jira/browse/KAFKA-4024?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-4024: - Description: Recently I updated our KafkaProducer configuration, specifically we adjusted {{retry

[jira] [Updated] (KAFKA-4024) First metadata update always take retry.backoff.ms milliseconds to complete

[ https://issues.apache.org/jira/browse/KAFKA-4024?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-4024: - Description: gtgtRecently I updated our KafkaProducer configuration, specifically we adjusted {{r

[jira] [Commented] (KAFKA-4024) First metadata update always take retry.backoff.ms milliseconds to complete

[ https://issues.apache.org/jira/browse/KAFKA-4024?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15458639#comment-15458639 ] Yuto Kawamura commented on KAFKA-4024: -- I reconsidered this issue and think I found t

[jira] [Commented] (KAFKA-4024) First metadata update always take retry.backoff.ms milliseconds to complete

[ https://issues.apache.org/jira/browse/KAFKA-4024?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15458699#comment-15458699 ] Yuto Kawamura commented on KAFKA-4024: -- Updated PR to fix this issue not only about t

[jira] [Assigned] (KAFKA-4116) Specifying 0.0.0.0 in "listeners" doesn't work

[ https://issues.apache.org/jira/browse/KAFKA-4116?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura reassigned KAFKA-4116: Assignee: Gwen Shapira (was: Yuto Kawamura) > Specifying 0.0.0.0 in "listeners" doesn't wo

[jira] [Updated] (KAFKA-4116) Specifying 0.0.0.0 in "listeners" doesn't work

[ https://issues.apache.org/jira/browse/KAFKA-4116?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuto Kawamura updated KAFKA-4116: - Assignee: Yuto Kawamura (was: Gwen Shapira) > Specifying 0.0.0.0 in "listeners" doesn't work > --

[jira] [Comment Edited] (KAFKA-4024) First metadata update always take retry.backoff.ms milliseconds to complete

[ https://issues.apache.org/jira/browse/KAFKA-4024?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15458639#comment-15458639 ] Yuto Kawamura edited comment on KAFKA-4024 at 9/7/16 1:46 AM: --