[jira] [Commented] (FLINK-16428) Fine-grained network buffer management for backpressure

[ https://issues.apache.org/jira/browse/FLINK-16428?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17400195#comment-17400195 ] Yingjie Cao commented on FLINK-16428: - [~joemoe] I will test this. > Fine-grained network buffer

[jira] [Commented] (FLINK-23724) Network buffer leak when ResultPartition is released (failover)

[ https://issues.apache.org/jira/browse/FLINK-23724?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17397807#comment-17397807 ] Yingjie Cao commented on FLINK-23724: -  [~pnowojski] [~akalashnikov], do you have time to take a

[jira] [Updated] (FLINK-23724) Network buffer leak when ResultPartition is released (failover)

[ https://issues.apache.org/jira/browse/FLINK-23724?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-23724: Affects Version/s: 1.13.2 > Network buffer leak when ResultPartition is released (failover) >

[jira] [Updated] (FLINK-23724) Network buffer leak when ResultPartition is released (failover)

[ https://issues.apache.org/jira/browse/FLINK-23724?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-23724: Fix Version/s: 1.13.3 > Network buffer leak when ResultPartition is released (failover) >

[jira] [Created] (FLINK-23724) Network buffer leak when ResultPartition is released (failover)

Yingjie Cao created FLINK-23724: --- Summary: Network buffer leak when ResultPartition is released (failover) Key: FLINK-23724 URL: https://issues.apache.org/jira/browse/FLINK-23724 Project: Flink

[jira] [Closed] (FLINK-22910) Refine ShuffleMaster lifecycle management for pluggable shuffle service framework

[ https://issues.apache.org/jira/browse/FLINK-22910?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao closed FLINK-22910. --- Release Note: We improved the ShuffleMaster interface by adding some lifecycle methods, including

[jira] [Commented] (FLINK-22910) Refine ShuffleMaster lifecycle management for pluggable shuffle service framework

[ https://issues.apache.org/jira/browse/FLINK-22910?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17394648#comment-17394648 ] Yingjie Cao commented on FLINK-22910: - Resolved via 

[jira] [Updated] (FLINK-23275) Support to release cluster partitions stored externally

[ https://issues.apache.org/jira/browse/FLINK-23275?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-23275: Parent: FLINK-23586 Issue Type: Sub-task (was: Improvement) > Support to release cluster

[jira] [Updated] (FLINK-23275) Support to release cluster partitions stored externally

[ https://issues.apache.org/jira/browse/FLINK-23275?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-23275: Parent: (was: FLINK-22672) Issue Type: Improvement (was: Sub-task) > Support to

[jira] [Updated] (FLINK-21788) Throw PartitionNotFoundException if the partition file has been lost for blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-21788?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21788: Labels: pull-request-available (was: pull-request-available stale-assigned) > Throw

[jira] [Commented] (FLINK-23382) Performance regression on 13.07.2021

[ https://issues.apache.org/jira/browse/FLINK-23382?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17387000#comment-17387000 ] Yingjie Cao commented on FLINK-23382: - [~pnowojski] Thanks for doing the benchmark. I also ran

[jira] [Updated] (FLINK-23214) Make ShuffleMaster a cluster level shared service

[ https://issues.apache.org/jira/browse/FLINK-23214?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-23214: Parent: FLINK-22910 Issue Type: Sub-task (was: Improvement) > Make ShuffleMaster a

[jira] [Updated] (FLINK-23214) Make ShuffleMaster a cluster level shared service

[ https://issues.apache.org/jira/browse/FLINK-23214?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-23214: Parent: (was: FLINK-22672) Issue Type: Improvement (was: Sub-task) > Make

[jira] [Updated] (FLINK-22674) Provide JobID when apply shuffle resource by ShuffleMaster

[ https://issues.apache.org/jira/browse/FLINK-22674?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22674: Parent: FLINK-22910 Issue Type: Sub-task (was: Improvement) > Provide JobID when apply

[jira] [Updated] (FLINK-22674) Provide JobID when apply shuffle resource by ShuffleMaster

[ https://issues.apache.org/jira/browse/FLINK-22674?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22674: Parent: (was: FLINK-22672) Issue Type: Improvement (was: Sub-task) > Provide JobID

[jira] [Updated] (FLINK-23249) Introduce ShuffleMasterContext to ShuffleMaster

[ https://issues.apache.org/jira/browse/FLINK-23249?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-23249: Parent: FLINK-22910 Issue Type: Sub-task (was: Improvement) > Introduce

[jira] [Updated] (FLINK-23249) Introduce ShuffleMasterContext to ShuffleMaster

[ https://issues.apache.org/jira/browse/FLINK-23249?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-23249: Parent: (was: FLINK-22672) Issue Type: Improvement (was: Sub-task) > Introduce

[jira] [Updated] (FLINK-22675) Add lifecycle methods to ShuffleMaster

[ https://issues.apache.org/jira/browse/FLINK-22675?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22675: Parent: FLINK-22910 Issue Type: Sub-task (was: Improvement) > Add lifecycle methods to

[jira] [Updated] (FLINK-22675) Add lifecycle methods to ShuffleMaster

[ https://issues.apache.org/jira/browse/FLINK-22675?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22675: Parent: (was: FLINK-22672) Issue Type: Improvement (was: Sub-task) > Add lifecycle

[jira] [Updated] (FLINK-22910) Refine ShuffleMaster lifecycle management for pluggable shuffle service framework

[ https://issues.apache.org/jira/browse/FLINK-22910?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22910: Parent: (was: FLINK-22672) Issue Type: Improvement (was: Sub-task) > Refine

[jira] [Commented] (FLINK-23382) Performance regression on 13.07.2021

[ https://issues.apache.org/jira/browse/FLINK-23382?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17380556#comment-17380556 ] Yingjie Cao commented on FLINK-23382: - [~pnowojski] Thanks for reporting this, I will take a look.

[jira] [Commented] (FLINK-16012) Reduce the default number of exclusive buffers from 2 to 1 on receiver side

[ https://issues.apache.org/jira/browse/FLINK-16012?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17379748#comment-17379748 ] Yingjie Cao commented on FLINK-16012: - Thanks [~pnowojski]. Looking forward to FLIP-183. > Reduce

[jira] [Commented] (FLINK-16012) Reduce the default number of exclusive buffers from 2 to 1 on receiver side

[ https://issues.apache.org/jira/browse/FLINK-16012?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17379549#comment-17379549 ] Yingjie Cao commented on FLINK-16012: - Hi [~pnowojski], do we still need this? I am asking because

[jira] [Updated] (FLINK-22675) Add lifecycle methods to ShuffleMaster

[ https://issues.apache.org/jira/browse/FLINK-22675?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22675: Summary: Add lifecycle methods to ShuffleMaster (was: Add lifecycle method to ShuffleMaster) >

[jira] [Updated] (FLINK-22675) Add lifecycle method to ShuffleMaster

[ https://issues.apache.org/jira/browse/FLINK-22675?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22675: Summary: Add lifecycle method to ShuffleMaster (was: Add an interface method

[jira] [Updated] (FLINK-22910) Refine ShuffleMaster lifecycle management for pluggable shuffle service framework

[ https://issues.apache.org/jira/browse/FLINK-22910?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22910: Summary: Refine ShuffleMaster lifecycle management for pluggable shuffle service framework (was:

[jira] [Commented] (FLINK-22672) Some enhancements for pluggable shuffle service framework

[ https://issues.apache.org/jira/browse/FLINK-22672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17378955#comment-17378955 ] Yingjie Cao commented on FLINK-22672: - [~maguowei] Thanks for the suggestion, I totally agree with

[jira] [Updated] (FLINK-23275) Support to release cluster partitions stored externally

[ https://issues.apache.org/jira/browse/FLINK-23275?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-23275: Fix Version/s: (was: 1.14.0) > Support to release cluster partitions stored externally >

[jira] [Created] (FLINK-23275) Support to release cluster partitions stored externally

Yingjie Cao created FLINK-23275: --- Summary: Support to release cluster partitions stored externally Key: FLINK-23275 URL: https://issues.apache.org/jira/browse/FLINK-23275 Project: Flink Issue

[jira] [Commented] (FLINK-23235) SinkITCase.writerAndCommitterAndGlobalCommitterExecuteInStreamingMode fails on azure

[ https://issues.apache.org/jira/browse/FLINK-23235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17375128#comment-17375128 ] Yingjie Cao commented on FLINK-23235: - The error seems to be similar. But FLINK-20010 has been

[jira] [Created] (FLINK-23249) Introduce ShuffleMasterContext to ShuffleMaster

Yingjie Cao created FLINK-23249: --- Summary: Introduce ShuffleMasterContext to ShuffleMaster Key: FLINK-23249 URL: https://issues.apache.org/jira/browse/FLINK-23249 Project: Flink Issue Type:

[jira] [Updated] (FLINK-23214) Make ShuffleMaster a cluster level shared service

[ https://issues.apache.org/jira/browse/FLINK-23214?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-23214: Component/s: Runtime / Coordination > Make ShuffleMaster a cluster level shared service >

[jira] [Created] (FLINK-23214) Make ShuffleMaster a cluster level shared service

Yingjie Cao created FLINK-23214: --- Summary: Make ShuffleMaster a cluster level shared service Key: FLINK-23214 URL: https://issues.apache.org/jira/browse/FLINK-23214 Project: Flink Issue Type:

[jira] [Updated] (FLINK-16012) Reduce the default number of exclusive buffers from 2 to 1 on receiver side

[ https://issues.apache.org/jira/browse/FLINK-16012?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-16012: Labels: pull-request-available (was: pull-request-available stale-assigned) > Reduce the default

[jira] [Updated] (FLINK-16428) Fine-grained network buffer management for backpressure

[ https://issues.apache.org/jira/browse/FLINK-16428?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-16428: Labels: (was: stale-critical) > Fine-grained network buffer management for backpressure >

[jira] [Updated] (FLINK-15024) System classloader memory leak after loading too many codegen classes.

[ https://issues.apache.org/jira/browse/FLINK-15024?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-15024: Labels: (was: stale-major) > System classloader memory leak after loading too many codegen

[jira] [Updated] (FLINK-21788) Throw PartitionNotFoundException if the partition file has been lost for blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-21788?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21788: Labels: pull-request-available (was: pull-request-available stale-assigned) > Throw

[jira] [Updated] (FLINK-15455) Enable TCP connection reuse across multiple jobs.

[ https://issues.apache.org/jira/browse/FLINK-15455?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-15455: Labels: (was: stale-major) > Enable TCP connection reuse across multiple jobs. >

[jira] [Updated] (FLINK-16641) Announce sender's backlog to solve the deadlock issue without exclusive buffers

[ https://issues.apache.org/jira/browse/FLINK-16641?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-16641: Labels: pull-request-available (was: pull-request-available stale-assigned) > Announce sender's

[jira] [Updated] (FLINK-16428) Fine-grained network buffer management for backpressure

[ https://issues.apache.org/jira/browse/FLINK-16428?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-16428: Labels: (was: stale-critical) > Fine-grained network buffer management for backpressure >

[jira] [Updated] (FLINK-16012) Reduce the default number of exclusive buffers from 2 to 1 on receiver side

[ https://issues.apache.org/jira/browse/FLINK-16012?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-16012: Labels: pull-request-available (was: pull-request-available stale-assigned) > Reduce the default

[jira] [Comment Edited] (FLINK-22910) ShuffleMaster enhancement for pluggable shuffle service framework

[ https://issues.apache.org/jira/browse/FLINK-22910?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17359026#comment-17359026 ] Yingjie Cao edited comment on FLINK-22910 at 6/8/21, 4:15 AM: -- This ticket

[jira] [Commented] (FLINK-22910) ShuffleMaster enhancement for pluggable shuffle service framework

[ https://issues.apache.org/jira/browse/FLINK-22910?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17359026#comment-17359026 ] Yingjie Cao commented on FLINK-22910: - This ticket cover FLINK-22674 and FLINK-22674. >

[jira] [Comment Edited] (FLINK-22910) ShuffleMaster enhancement for pluggable shuffle service framework

[ https://issues.apache.org/jira/browse/FLINK-22910?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17359026#comment-17359026 ] Yingjie Cao edited comment on FLINK-22910 at 6/8/21, 4:05 AM: -- This ticket

[jira] [Created] (FLINK-22910) ShuffleMaster enhancement for pluggable shuffle service framework

Yingjie Cao created FLINK-22910: --- Summary: ShuffleMaster enhancement for pluggable shuffle service framework Key: FLINK-22910 URL: https://issues.apache.org/jira/browse/FLINK-22910 Project: Flink

[jira] [Updated] (FLINK-13203) [proper fix] Deadlock occurs when requiring exclusive buffer for RemoteInputChannel

[ https://issues.apache.org/jira/browse/FLINK-13203?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-13203: Labels: auto-deprioritized-critical (was: auto-deprioritized-critical stale-major) > [proper

[jira] [Updated] (FLINK-21788) Throw PartitionNotFoundException if the partition file has been lost for blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-21788?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21788: Labels: pull-request-available (was: pull-request-available stale-assigned) > Throw

[jira] [Commented] (FLINK-16641) Announce sender's backlog to solve the deadlock issue without exclusive buffers

[ https://issues.apache.org/jira/browse/FLINK-16641?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17354938#comment-17354938 ] Yingjie Cao commented on FLINK-16641: - [~pnowojski] Thanks very much. If there is anything I can

[jira] [Updated] (FLINK-16012) Reduce the default number of exclusive buffers from 2 to 1 on receiver side

[ https://issues.apache.org/jira/browse/FLINK-16012?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-16012: Labels: pull-request-available (was: pull-request-available stale-assigned) > Reduce the default

[jira] [Updated] (FLINK-15024) System classloader memory leak after loading too many codegen classes.

[ https://issues.apache.org/jira/browse/FLINK-15024?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-15024: Priority: Major (was: Minor) > System classloader memory leak after loading too many codegen

[jira] [Updated] (FLINK-15024) System classloader memory leak after loading too many codegen classes.

[ https://issues.apache.org/jira/browse/FLINK-15024?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-15024: Labels: (was: auto-deprioritized-major) > System classloader memory leak after loading too many

[jira] [Updated] (FLINK-16428) Fine-grained network buffer management for backpressure

[ https://issues.apache.org/jira/browse/FLINK-16428?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-16428: Labels: (was: stale-critical) > Fine-grained network buffer management for backpressure >

[jira] [Updated] (FLINK-21788) Throw PartitionNotFoundException if the partition file has been lost for blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-21788?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21788: Labels: pull-request-available (was: pull-request-available stale-assigned) > Throw

[jira] [Commented] (FLINK-22643) Too many TCP connections among TaskManagers for large scale jobs

[ https://issues.apache.org/jira/browse/FLINK-22643?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17345907#comment-17345907 ] Yingjie Cao commented on FLINK-22643: - I think it is also important for ad-hoc query scenario. When

[jira] [Commented] (FLINK-16641) Announce sender's backlog to solve the deadlock issue without exclusive buffers

[ https://issues.apache.org/jira/browse/FLINK-16641?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17342977#comment-17342977 ] Yingjie Cao commented on FLINK-16641: - Hi [~pnowojski], I have rebased the PR on the latest master

[jira] [Closed] (FLINK-18762) Make network buffers per incoming/outgoing channel can be configured separately

[ https://issues.apache.org/jira/browse/FLINK-18762?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao closed FLINK-18762. --- Resolution: Won't Fix > Make network buffers per incoming/outgoing channel can be configured >

[jira] [Commented] (FLINK-18762) Make network buffers per incoming/outgoing channel can be configured separately

[ https://issues.apache.org/jira/browse/FLINK-18762?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17341214#comment-17341214 ] Yingjie Cao commented on FLINK-18762: - This ticket is not needed, so closing it. > Make network

[jira] [Updated] (FLINK-16428) Fine-grained network buffer management for backpressure

[ https://issues.apache.org/jira/browse/FLINK-16428?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-16428: Labels: (was: stale-critical) > Fine-grained network buffer management for backpressure >

[jira] [Updated] (FLINK-15455) Enable TCP connection reuse across multiple jobs.

[ https://issues.apache.org/jira/browse/FLINK-15455?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-15455: Labels: (was: stale-major) > Enable TCP connection reuse across multiple jobs. >

[jira] [Commented] (FLINK-22401) Python tests fail with "Slot ... is not allocated by job ..."

[ https://issues.apache.org/jira/browse/FLINK-22401?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17329961#comment-17329961 ] Yingjie Cao commented on FLINK-22401: - [~trohrmann] [~xintongsong] Really thanks for the reply and

[jira] [Commented] (FLINK-22401) Python tests fail with "Slot ... is not allocated by job ..."

[ https://issues.apache.org/jira/browse/FLINK-22401?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17327232#comment-17327232 ] Yingjie Cao commented on FLINK-22401: - BTW, I noticed that the above config option was newly

[jira] [Commented] (FLINK-22401) Python tests fail with "Slot ... is not allocated by job ..."

[ https://issues.apache.org/jira/browse/FLINK-22401?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17327214#comment-17327214 ] Yingjie Cao commented on FLINK-22401: - Is the timeout controled by config option 

[jira] [Updated] (FLINK-21789) Make FileChannelManagerImpl#getNextPathNum select data directories fairly

[ https://issues.apache.org/jira/browse/FLINK-21789?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21789: Fix Version/s: (was: 1.11.4) 1.14.0 > Make

[jira] [Updated] (FLINK-21790) Shuffle data directories to make directory selection of different TaskManagers fairer

[ https://issues.apache.org/jira/browse/FLINK-21790?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21790: Fix Version/s: (was: 1.13.0) 1.14.0 > Shuffle data directories to make

[jira] [Updated] (FLINK-21789) Make FileChannelManagerImpl#getNextPathNum select data directories fairly

[ https://issues.apache.org/jira/browse/FLINK-21789?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21789: Fix Version/s: (was: 1.13.0) 1.11.4 > Make

[jira] [Commented] (FLINK-22385) Type mismatch in NetworkBufferPool

[ https://issues.apache.org/jira/browse/FLINK-22385?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17326244#comment-17326244 ] Yingjie Cao commented on FLINK-22385: - cc [~pnowojski]. > Type mismatch in NetworkBufferPool >

[jira] [Closed] (FLINK-19614) Further optimization of sort-merge based blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-19614?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao closed FLINK-19614. --- Resolution: Resolved > Further optimization of sort-merge based blocking shuffle >

[jira] [Closed] (FLINK-22307) Increase the default value of data writing cache size (not configurable) for sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22307?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao closed FLINK-22307. --- Resolution: Resolved > Increase the default value of data writing cache size (not configurable) for

[jira] [Commented] (FLINK-22307) Increase the default value of data writing cache size (not configurable) for sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22307?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17324396#comment-17324396 ] Yingjie Cao commented on FLINK-22307: - Merged via d29357791315ace0218308f9f96bf65c7c8079d7 on

[jira] [Commented] (FLINK-22305) Improve log messages of sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22305?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17324394#comment-17324394 ] Yingjie Cao commented on FLINK-22305: - Merged via 77e0478a7879bd41f9f52872d84c467d158b8098 on

[jira] [Closed] (FLINK-22305) Improve log messages of sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22305?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao closed FLINK-22305. --- Resolution: Resolved > Improve log messages of sort-merge blocking shuffle >

[jira] [Updated] (FLINK-22307) Increase the default value of data writing cache size (not configurable) for sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22307?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22307: Summary: Increase the default value of data writing cache size (not configurable) for sort-merge

[jira] [Updated] (FLINK-22307) Increase the default value of data writing cache size for sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22307?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22307: Summary: Increase the default value of data writing cache size for sort-merge blocking shuffle

[jira] [Updated] (FLINK-22305) Improve log messages of sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22305?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22305: Summary: Improve log messages of sort-merge blocking shuffle (was: Improve log message of

[jira] [Updated] (FLINK-22305) Improve log message of sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22305?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22305: Summary: Improve log message of sort-merge blocking shuffle (was: Increase the default value of

[jira] [Commented] (FLINK-22305) Increase the default value of taskmanager.network.sort-shuffle.min-buffers

[ https://issues.apache.org/jira/browse/FLINK-22305?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17324181#comment-17324181 ] Yingjie Cao commented on FLINK-22305: - After some tests, I find that increasing this default value

[jira] [Updated] (FLINK-16012) Reduce the default number of exclusive buffers from 2 to 1 on receiver side

[ https://issues.apache.org/jira/browse/FLINK-16012?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-16012: Labels: pull-request-available (was: pull-request-available stale-assigned) > Reduce the default

[jira] [Commented] (FLINK-16012) Reduce the default number of exclusive buffers from 2 to 1 on receiver side

[ https://issues.apache.org/jira/browse/FLINK-16012?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17324164#comment-17324164 ] Yingjie Cao commented on FLINK-16012: - I am still working on it. > Reduce the default number of

[jira] [Updated] (FLINK-16641) Announce sender's backlog to solve the deadlock issue without exclusive buffers

[ https://issues.apache.org/jira/browse/FLINK-16641?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-16641: Labels: pull-request-available (was: pull-request-available stale-assigned) > Announce sender's

[jira] [Commented] (FLINK-21788) Throw PartitionNotFoundException if the partition file has been lost for blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-21788?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17324161#comment-17324161 ] Yingjie Cao commented on FLINK-21788: - I am still working on it. > Throw PartitionNotFoundException

[jira] [Commented] (FLINK-16641) Announce sender's backlog to solve the deadlock issue without exclusive buffers

[ https://issues.apache.org/jira/browse/FLINK-16641?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17324162#comment-17324162 ] Yingjie Cao commented on FLINK-16641: - I am still working on it. > Announce sender's backlog to

[jira] [Updated] (FLINK-21788) Throw PartitionNotFoundException if the partition file has been lost for blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-21788?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21788: Labels: pull-request-available (was: pull-request-available stale-assigned) > Throw

[jira] [Created] (FLINK-22307) Increase the data writing cache size of sort-merge blocking shuffle

Yingjie Cao created FLINK-22307: --- Summary: Increase the data writing cache size of sort-merge blocking shuffle Key: FLINK-22307 URL: https://issues.apache.org/jira/browse/FLINK-22307 Project: Flink

[jira] [Created] (FLINK-22305) Increase the default value of taskmanager.network.sort-shuffle.min-buffers

Yingjie Cao created FLINK-22305: --- Summary: Increase the default value of taskmanager.network.sort-shuffle.min-buffers Key: FLINK-22305 URL: https://issues.apache.org/jira/browse/FLINK-22305 Project:

[jira] [Resolved] (FLINK-22153) Manually test the sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22153?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao resolved FLINK-22153. - Resolution: Resolved > Manually test the sort-merge blocking shuffle >

[jira] [Commented] (FLINK-22153) Manually test the sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22153?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17322707#comment-17322707 ] Yingjie Cao commented on FLINK-22153: - OK, I will create a follow up ticket. > Manually test the

[jira] [Commented] (FLINK-22153) Manually test the sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22153?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17322641#comment-17322641 ] Yingjie Cao commented on FLINK-22153: - [~trohrmann] Thanks for testing the sort-merge blocking

[jira] [Commented] (FLINK-21859) Batch job fails due to "Could not mark slot 61a637e3977c58a0e6b73533c419297d active"

[ https://issues.apache.org/jira/browse/FLINK-21859?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17320169#comment-17320169 ] Yingjie Cao commented on FLINK-21859: - [~trohrmann] I have updated the log with DEBUG enabled. >

[jira] [Updated] (FLINK-21859) Batch job fails due to "Could not mark slot 61a637e3977c58a0e6b73533c419297d active"

[ https://issues.apache.org/jira/browse/FLINK-21859?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21859: Attachment: jm.log.zip > Batch job fails due to "Could not mark slot

[jira] [Updated] (FLINK-21859) Batch job fails due to "Could not mark slot 61a637e3977c58a0e6b73533c419297d active"

[ https://issues.apache.org/jira/browse/FLINK-21859?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21859: Attachment: tm1.log.zip > Batch job fails due to "Could not mark slot

[jira] [Updated] (FLINK-21859) Batch job fails due to "Could not mark slot 61a637e3977c58a0e6b73533c419297d active"

[ https://issues.apache.org/jira/browse/FLINK-21859?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21859: Attachment: tm2.log.zip > Batch job fails due to "Could not mark slot

[jira] [Updated] (FLINK-21967) Add documentation on the operation of blocking result partition

[ https://issues.apache.org/jira/browse/FLINK-21967?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21967: Priority: Critical (was: Blocker) > Add documentation on the operation of blocking result

[jira] [Commented] (FLINK-21859) Batch job fails due to "Could not mark slot 61a637e3977c58a0e6b73533c419297d active"

[ https://issues.apache.org/jira/browse/FLINK-21859?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17319083#comment-17319083 ] Yingjie Cao commented on FLINK-21859: - OK, I will try to reproduce the issue with DEBUG logs

[jira] [Updated] (FLINK-21967) Add documentation on the operation of blocking result partition

[ https://issues.apache.org/jira/browse/FLINK-21967?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21967: Priority: Blocker (was: Major) > Add documentation on the operation of blocking result partition

[jira] [Created] (FLINK-22156) HiveDialectQueryITCase fails on Azure because of no output for 900 seconds

Yingjie Cao created FLINK-22156: --- Summary: HiveDialectQueryITCase fails on Azure because of no output for 900 seconds Key: FLINK-22156 URL: https://issues.apache.org/jira/browse/FLINK-22156 Project:

[jira] [Created] (FLINK-22153) Manually test the sort-merge blocking shuffle

Yingjie Cao created FLINK-22153: --- Summary: Manually test the sort-merge blocking shuffle Key: FLINK-22153 URL: https://issues.apache.org/jira/browse/FLINK-22153 Project: Flink Issue Type: Task

[jira] [Updated] (FLINK-22153) Manually test the sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-22153?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-22153: Priority: Major (was: Blocker) > Manually test the sort-merge blocking shuffle >

[jira] [Updated] (FLINK-21850) Improve document and config description of sort-merge blocking shuffle

[ https://issues.apache.org/jira/browse/FLINK-21850?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yingjie Cao updated FLINK-21850: Priority: Blocker (was: Major) > Improve document and config description of sort-merge blocking

[jira] [Created] (FLINK-22127) Enrich error message of read buffer request timeout exception

Yingjie Cao created FLINK-22127: --- Summary: Enrich error message of read buffer request timeout exception Key: FLINK-22127 URL: https://issues.apache.org/jira/browse/FLINK-22127 Project: Flink

[jira] [Commented] (FLINK-21859) Batch job fails due to "Could not mark slot 61a637e3977c58a0e6b73533c419297d active"

[ https://issues.apache.org/jira/browse/FLINK-21859?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17312166#comment-17312166 ] Yingjie Cao commented on FLINK-21859: - tm log updated, the previous one is not right. > Batch job

<    1   2   3   4   5   6   7   8   >