[jira] [Commented] (FLINK-12203) Refactor ResultPartitionManager to break tie with Task
[ https://issues.apache.org/jira/browse/FLINK-12203?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16829273#comment-16829273 ] Andrey Zagrebin commented on FLINK-12203: - Hey [~rmetzger], thanks for correction. I will try to keep components up to date. > Refactor ResultPartitionManager to break tie with Task > -- > > Key: FLINK-12203 > URL: https://issues.apache.org/jira/browse/FLINK-12203 > Project: Flink > Issue Type: Sub-task > Components: Runtime / Network >Reporter: Andrey Zagrebin >Assignee: Andrey Zagrebin >Priority: Major > Labels: pull-request-available > Time Spent: 10m > Remaining Estimate: 0h > > At the moment, we have ResultPartitionManager.releasePartitionsProducedBy > which uses indexing by task in network environment. These methods are > eventually used only by Task which already knows its partitions so Task can > use ResultPartition.fail(cause) and TaskExecutor.failPartition could directly > use NetworkEnviroment.releasePartitions(Collection). This > also requires that JM Execution sends produced partition ids instead of just > ExecutionAttemptID. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (FLINK-12203) Refactor ResultPartitionManager to break tie with Task
[ https://issues.apache.org/jira/browse/FLINK-12203?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16829239#comment-16829239 ] Robert Metzger commented on FLINK-12203: Hey [~azagrebin], thanks a lot for opening a ticket in the FLINK bug tracker. I just manually assigned the ticket to a component. For future tickets, please remember to always assign a new issue to a component, so that the component owner can pick it up. > Refactor ResultPartitionManager to break tie with Task > -- > > Key: FLINK-12203 > URL: https://issues.apache.org/jira/browse/FLINK-12203 > Project: Flink > Issue Type: Sub-task > Components: Runtime / Network >Reporter: Andrey Zagrebin >Assignee: Andrey Zagrebin >Priority: Major > Labels: pull-request-available > Time Spent: 10m > Remaining Estimate: 0h > > At the moment, we have ResultPartitionManager.releasePartitionsProducedBy > which uses indexing by task in network environment. These methods are > eventually used only by Task which already knows its partitions so Task can > use ResultPartition.fail(cause) and TaskExecutor.failPartition could directly > use NetworkEnviroment.releasePartitions(Collection). This > also requires that JM Execution sends produced partition ids instead of just > ExecutionAttemptID. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
