[
https://issues.apache.org/jira/browse/FLINK-13100?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16880465#comment-16880465
]
zhijiang commented on FLINK-13100:
--
Yes, the previous SpilledSubpartitionView really has the deadlock
[
https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16880427#comment-16880427
]
zhijiang commented on FLINK-11082:
--
[~pnowojski], if the requested floating buffers are not used in
zhijiang created FLINK-13141:
Summary: Remove getBufferSize method from BufferPoolFactory
Key: FLINK-13141
URL: https://issues.apache.org/jira/browse/FLINK-13141
Project: Flink
Issue Type:
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12852:
-
Comment: was deleted
(was: I thought of another option to keep the dynamic local pool as now and
only
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16879997#comment-16879997
]
zhijiang commented on FLINK-12852:
--
I thought of another option to keep the dynamic local pool as now
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16879396#comment-16879396
]
zhijiang edited comment on FLINK-12852 at 7/5/19 3:58 PM:
--
Thanks for above
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16879396#comment-16879396
]
zhijiang commented on FLINK-12852:
--
Thanks for above detail conclusions [~StephanEwen]!
I agree that
zhijiang created FLINK-13126:
Summary: Construct special test/benchmark to verify the backlog
effect
Key: FLINK-13126
URL: https://issues.apache.org/jira/browse/FLINK-13126
Project: Flink
Issue
[
https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-11082:
-
Affects Version/s: (was: 1.6.3)
(was: 1.8.0)
> Fix the calculation of
[
https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-11082:
-
Description:
The backlog of subpartition should indicate how many buffers are consumable,
then the
[
https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-11082:
-
Summary: Fix the calculation of backlog in PipelinedSubpartition (was:
Increase backlog only if it is
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16878692#comment-16878692
]
zhijiang commented on FLINK-12852:
--
Thanks for concerning this issue [~aitozi]
We already have the
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16878518#comment-16878518
]
zhijiang commented on FLINK-12852:
--
We have two directions in general: buffers sharing between tasks or
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16878516#comment-16878516
]
zhijiang commented on FLINK-12852:
--
[~pnowojski] "Network resource matching in slot" refers to the
[
https://issues.apache.org/jira/browse/FLINK-13100?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16878513#comment-16878513
]
zhijiang commented on FLINK-13100:
--
What do you think [~pnowojski] [~StephanEwen]
> Fix the unexpected
zhijiang created FLINK-13100:
Summary: Fix the unexpected IOException during
FileBufferReader#nextBuffer
Key: FLINK-13100
URL: https://issues.apache.org/jira/browse/FLINK-13100
Project: Flink
[
https://issues.apache.org/jira/browse/FLINK-12738?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang resolved FLINK-12738.
--
Resolution: Fixed
Fixed in b6563e385263beb556ce19855ca6dfdbb7f6c853
> Remove abstract getPageSize
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16878289#comment-16878289
]
zhijiang edited comment on FLINK-12852 at 7/4/19 3:27 AM:
--
Network resource
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16878289#comment-16878289
]
zhijiang commented on FLINK-12852:
--
Network resource matching in slot has many unsetting issues which
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16877737#comment-16877737
]
zhijiang commented on FLINK-12852:
--
It is hardly to say whether we could get performance benefits via
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16877616#comment-16877616
]
zhijiang commented on FLINK-12852:
--
Hey [~gaoyunhaii], considering the second point for long run, I
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16877605#comment-16877605
]
zhijiang commented on FLINK-12852:
--
The spill way could solve the deadlock issue in function, which is
[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16877073#comment-16877073
]
zhijiang commented on FLINK-12852:
--
[~StephanEwen] thanks for concerning this issue and sharing the
[
https://issues.apache.org/jira/browse/FLINK-12882?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12882:
-
Description: The ResultPartitionID could be got directly from
ResultPartitionDeploymentDescriptor, so
[
https://issues.apache.org/jira/browse/FLINK-12882?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12882:
-
Summary: Remove ExecutionAttemptID argument from
ResultPartitionFactory#create (was: Remove
zhijiang created FLINK-13010:
Summary: Refactor the process of SchedulerNG#requestPartitionState
Key: FLINK-13010
URL: https://issues.apache.org/jira/browse/FLINK-13010
Project: Flink
Issue
[
https://issues.apache.org/jira/browse/FLINK-12284?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12284:
-
Component/s: Runtime / Metrics
> InputBufferPoolUsage is incorrect in credit-based network control flow
[
https://issues.apache.org/jira/browse/FLINK-12889?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16872933#comment-16872933
]
zhijiang edited comment on FLINK-12889 at 6/26/19 5:33 AM:
---
My previous
[
https://issues.apache.org/jira/browse/FLINK-12889?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16872933#comment-16872933
]
zhijiang edited comment on FLINK-12889 at 6/26/19 5:32 AM:
---
My previous
[
https://issues.apache.org/jira/browse/FLINK-12889?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12889:
-
Priority: Minor (was: Blocker)
> Job keeps in FAILING state
> --
>
>
[
https://issues.apache.org/jira/browse/FLINK-12889?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12889:
-
Affects Version/s: (was: 1.7.2)
> Job keeps in FAILING state
> --
>
>
[
https://issues.apache.org/jira/browse/FLINK-12889?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16872933#comment-16872933
]
zhijiang commented on FLINK-12889:
--
My previous analysis in ML was as follows:
Actually all the five
[
https://issues.apache.org/jira/browse/FLINK-12889?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12889:
-
Component/s: (was: API / DataStream)
Runtime / Coordination
> Job keeps in FAILING
[
https://issues.apache.org/jira/browse/FLINK-12909?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16871942#comment-16871942
]
zhijiang commented on FLINK-12909:
--
Also you should connect the PR #7191 to this Jira ticket
[
https://issues.apache.org/jira/browse/FLINK-12909?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16871941#comment-16871941
]
zhijiang commented on FLINK-12909:
--
The ticket does not describes the issue well.
Any IOException
zhijiang created FLINK-12882:
Summary: Remove ExecutionAttemptID field from
ShuffleEnvironment#createResultPartitionWriters
Key: FLINK-12882
URL: https://issues.apache.org/jira/browse/FLINK-12882
[
https://issues.apache.org/jira/browse/FLINK-12882?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12882:
-
Description:
The {{ExecutionAttemptID}} is only used for constructing {{ResultPartitionID}}
during
[
https://issues.apache.org/jira/browse/FLINK-12843?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12843:
-
Description:
The pin logic is for adding the reference counter based on number of
subpartitions for
[
https://issues.apache.org/jira/browse/FLINK-12843?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12843:
-
Summary: Refactor the pin logic in ReleaseOnConsumptionResultPartition
(was: Refactor the pin logic in
[
https://issues.apache.org/jira/browse/FLINK-12070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16863947#comment-16863947
]
zhijiang edited comment on FLINK-12070 at 6/14/19 11:09 AM:
In Yingjie's
[
https://issues.apache.org/jira/browse/FLINK-12070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16863947#comment-16863947
]
zhijiang commented on FLINK-12070:
--
In Yingjie's testing based on session mode, the container is
zhijiang created FLINK-12843:
Summary: Refactor the pin logic in ResultPartition
Key: FLINK-12843
URL: https://issues.apache.org/jira/browse/FLINK-12843
Project: Flink
Issue Type: Sub-task
zhijiang created FLINK-12842:
Summary: Fix invalid check released state during
ResultPartition#createSubpartitionView
Key: FLINK-12842
URL: https://issues.apache.org/jira/browse/FLINK-12842
Project:
[
https://issues.apache.org/jira/browse/FLINK-12761?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12761:
-
Summary: Fine grained resource management (was: Fine grained resource
managedment)
> Fine grained
[
https://issues.apache.org/jira/browse/FLINK-12738?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12738:
-
Summary: Remove abstract getPageSize method from InputGate (was: Remove
abstract getPageSize from
[
https://issues.apache.org/jira/browse/FLINK-12740?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16856643#comment-16856643
]
zhijiang commented on FLINK-12740:
--
Yes, the new added uint test for FLINK-12544 might be not stable.
[
https://issues.apache.org/jira/browse/FLINK-12740?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang reassigned FLINK-12740:
Assignee: zhijiang
> SpillableSubpartitionTest deadlocks on Travis
>
zhijiang created FLINK-12738:
Summary: Remove abstract getPageSize from InputGate
Key: FLINK-12738
URL: https://issues.apache.org/jira/browse/FLINK-12738
Project: Flink
Issue Type: Sub-task
zhijiang created FLINK-12735:
Summary: Make shuffle environment implementation independent with
IOManager
Key: FLINK-12735
URL: https://issues.apache.org/jira/browse/FLINK-12735
Project: Flink
[
https://issues.apache.org/jira/browse/FLINK-11604?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang closed FLINK-11604.
Resolution: Duplicate
Already done in FLINK-12570
> Extend the necessary methods in
[
https://issues.apache.org/jira/browse/FLINK-12544?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16855696#comment-16855696
]
zhijiang commented on FLINK-12544:
--
Merged into release-1.8 : 61fa1005adc61b6fc3439fdd127e5f25adea
[
https://issues.apache.org/jira/browse/FLINK-12544?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang closed FLINK-12544.
Resolution: Fixed
Fix Version/s: 1.8.1
> Deadlock while releasing memory and requesting segment
[
https://issues.apache.org/jira/browse/FLINK-12201?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=1684#comment-1684
]
zhijiang commented on FLINK-12201:
--
master : 3351ca11a366713328a6d0cb0e87100cb8c24200
> Introduce
[
https://issues.apache.org/jira/browse/FLINK-12201?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang closed FLINK-12201.
Resolution: Fixed
> Introduce InputGateWithMetrics in Task to increment numBytesIn metric
>
[
https://issues.apache.org/jira/browse/FLINK-12070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16855252#comment-16855252
]
zhijiang edited comment on FLINK-12070 at 6/4/19 2:59 AM:
--
Agree with Stephan
[
https://issues.apache.org/jira/browse/FLINK-12070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16855252#comment-16855252
]
zhijiang edited comment on FLINK-12070 at 6/4/19 2:58 AM:
--
Agree with Stephan
[
https://issues.apache.org/jira/browse/FLINK-12070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16855252#comment-16855252
]
zhijiang commented on FLINK-12070:
--
Agree with Stephan and Stefan's opinions.
We could add the batch
[
https://issues.apache.org/jira/browse/FLINK-12603?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12603:
-
Summary: Remove getOwningTaskName method from InputGate (was: Refactor
InputGate interface to remove
[
https://issues.apache.org/jira/browse/FLINK-12603?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12603:
-
Description:
In order to make abstract InputGate simple for extending new implementations in
shuffle
[
https://issues.apache.org/jira/browse/FLINK-12642?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16851462#comment-16851462
]
zhijiang commented on FLINK-12642:
--
Yes, that is the ideal way which we confirmed before to make
[
https://issues.apache.org/jira/browse/FLINK-12544?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12544:
-
Description:
It is reported by flink user, and the original jstack is as following:
{code:java}
//
[
https://issues.apache.org/jira/browse/FLINK-12544?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12544:
-
Summary: Deadlock while releasing memory and requesting segment concurrent
in SpillableSubpartition
[
https://issues.apache.org/jira/browse/FLINK-12642?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16849508#comment-16849508
]
zhijiang commented on FLINK-12642:
--
Thanks for reporting this [~Zentol]!
I think we could check
zhijiang created FLINK-12630:
Summary: Refactor abstract InputGate to general interface
Key: FLINK-12630
URL: https://issues.apache.org/jira/browse/FLINK-12630
Project: Flink
Issue Type:
[
https://issues.apache.org/jira/browse/FLINK-12201?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12201:
-
Description: Incrementing of numBytesIn metric in SingleInputGate does not
depend on shuffle service
[
https://issues.apache.org/jira/browse/FLINK-12201?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12201:
-
Summary: Introduce InputGateWithMetrics in Task to increment numBytesIn
metric (was: Introduce
[
https://issues.apache.org/jira/browse/FLINK-12603?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12603:
-
Description:
In order to make abstract InputGate simple for extending new implementations in
shuffle
[
https://issues.apache.org/jira/browse/FLINK-12603?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12603:
-
Summary: Refactor InputGate interface to remove unnecessary methods (was:
Remove getOwningTaskName
zhijiang created FLINK-12603:
Summary: Remove getOwningTaskName method from InputGate
Key: FLINK-12603
URL: https://issues.apache.org/jira/browse/FLINK-12603
Project: Flink
Issue Type: Sub-task
[
https://issues.apache.org/jira/browse/FLINK-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-6227:
Summary: Introduce the abstract PartitionException for downstream task
failure (was: Introduce the
[
https://issues.apache.org/jira/browse/FLINK-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-6227:
Description:
It is part of FLIP-1.
We define a new abstract special exception called
[
https://issues.apache.org/jira/browse/FLINK-12458?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12458:
-
Description: If the consumer can not establish a connection to remote task
executor, which might
[
https://issues.apache.org/jira/browse/FLINK-12458?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12458:
-
Summary: Introduce PartitionConnectionException for unreachable producer
(was: Throw
zhijiang created FLINK-12571:
Summary: Make NetworkEnvironment#start() return the binded data
port
Key: FLINK-12571
URL: https://issues.apache.org/jira/browse/FLINK-12571
Project: Flink
Issue
[
https://issues.apache.org/jira/browse/FLINK-10653?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16844642#comment-16844642
]
zhijiang commented on FLINK-10653:
--
Thanks for the concerns. [~prudenko]
Actually we ever also
zhijiang created FLINK-12564:
Summary: Remove getBufferProvider method from
ResultPartitionWriter interface
Key: FLINK-12564
URL: https://issues.apache.org/jira/browse/FLINK-12564
Project: Flink
zhijiang created FLINK-12544:
Summary: Deadlock during releasing memory in SpillableSubpartition
Key: FLINK-12544
URL: https://issues.apache.org/jira/browse/FLINK-12544
Project: Flink
Issue
[
https://issues.apache.org/jira/browse/FLINK-12202?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16839963#comment-16839963
]
zhijiang commented on FLINK-12202:
--
Hey [~azagrebin], as we confirmed before, this ticket is not needed
[
https://issues.apache.org/jira/browse/FLINK-12474?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang closed FLINK-12474.
Resolution: Invalid
> UnionInputGate should be notified when closing SingleInputGate by canceler
>
[
https://issues.apache.org/jira/browse/FLINK-12474?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16839298#comment-16839298
]
zhijiang commented on FLINK-12474:
--
Yes, you are right after double checking the codes.
Because the
[
https://issues.apache.org/jira/browse/FLINK-12474?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16839230#comment-16839230
]
zhijiang commented on FLINK-12474:
--
[~pnowojski] thanks for the confirmation!
Yes, the
[
https://issues.apache.org/jira/browse/FLINK-12497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-12497:
-
Description:
In current {{ConnectionManager#start(ResultPartitionProvider,
TaskEventDispatcher)}}, the
zhijiang created FLINK-12497:
Summary: Refactor the start method of ConnectionManager
Key: FLINK-12497
URL: https://issues.apache.org/jira/browse/FLINK-12497
Project: Flink
Issue Type: Sub-task
[
https://issues.apache.org/jira/browse/FLINK-11403?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-11403:
-
Summary: Introduce ResultPartitionWithConsumableNotifier in task for
notifying consumable result
[
https://issues.apache.org/jira/browse/FLINK-11403?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-11403:
-
Description:
The creation of {{ResultPartitionWriter}} from {{NetworkEnvironment}} relies on
[
https://issues.apache.org/jira/browse/FLINK-12474?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16836964#comment-16836964
]
zhijiang commented on FLINK-12474:
--
[~pnowojski], do you think this jira is worth fixing or my
zhijiang created FLINK-12474:
Summary: UnionInputGate should be notified when closing
SingleInputGate by canceler thread
Key: FLINK-12474
URL: https://issues.apache.org/jira/browse/FLINK-12474
Project:
zhijiang created FLINK-12458:
Summary: Throw PartitionNotFoundException if consumer can not
establish a connection to remote TM
Key: FLINK-12458
URL: https://issues.apache.org/jira/browse/FLINK-12458
zhijiang created FLINK-12331:
Summary: Introduce partition/gate setup to decouple task
registration with NetworkEnvironment
Key: FLINK-12331
URL: https://issues.apache.org/jira/browse/FLINK-12331
[
https://issues.apache.org/jira/browse/FLINK-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang updated FLINK-6227:
Component/s: Runtime / Coordination
> Introduce the DataConsumptionException for downstream task failure
>
zhijiang created FLINK-12213:
Summary: Pass TaskManagerMetricGroup into constructor of
NetworkEnvironment
Key: FLINK-12213
URL: https://issues.apache.org/jira/browse/FLINK-12213
Project: Flink
[
https://issues.apache.org/jira/browse/FLINK-12070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16818619#comment-16818619
]
zhijiang edited comment on FLINK-12070 at 4/16/19 4:25 AM:
---
Thanks for
[
https://issues.apache.org/jira/browse/FLINK-12070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16818619#comment-16818619
]
zhijiang edited comment on FLINK-12070 at 4/16/19 4:24 AM:
---
Thanks for
[
https://issues.apache.org/jira/browse/FLINK-12070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16818619#comment-16818619
]
zhijiang commented on FLINK-12070:
--
Thanks for providing the POC for this feature [~StephanEwen]
I
[
https://issues.apache.org/jira/browse/FLINK-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16818018#comment-16818018
]
zhijiang commented on FLINK-6227:
-
Ok, we would focus on the implementation soon.
Also we could fully
zhijiang created FLINK-12154:
Summary: Remove legacy fields for SingleInputGate
Key: FLINK-12154
URL: https://issues.apache.org/jira/browse/FLINK-12154
Project: Flink
Issue Type: Sub-task
zhijiang created FLINK-12146:
Summary: Remove unregister task from NetworkEnvironment to
simplify the interface of ShuffleService
Key: FLINK-12146
URL: https://issues.apache.org/jira/browse/FLINK-12146
[
https://issues.apache.org/jira/browse/FLINK-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16813215#comment-16813215
]
zhijiang commented on FLINK-6227:
-
Exactly, this exception is received on consumer side via network stack
zhijiang created FLINK-12127:
Summary: Move network related options to NetworkEnvironmentOptions
Key: FLINK-12127
URL: https://issues.apache.org/jira/browse/FLINK-12127
Project: Flink
Issue
[
https://issues.apache.org/jira/browse/FLINK-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhijiang reassigned FLINK-6227:
---
Assignee: zhijiang
> Introduce the DataConsumptionException for downstream task failure
>
801 - 900 of 1161 matches
Mail list logo
