[
https://issues.apache.org/jira/browse/FLINK-12876?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zhu Zhu updated FLINK-12876:
----------------------------
Description:
We need an adapter to adapt flip1.RestartPipelinedRegionStrategy for legacy
scheduler, so that the legacy scheduler can support fine grained recovery.
The failover recovery should respect scheduleMode(EAGER/LAZY_FROM_SOURCES) to
avoid more tasks get re-scheduled than that before failover.
was:We need a adapter to adapt flip1.RestartPipelinedRegionStrategy for
legacy scheduler, so that the legacy scheduler can support fine grained
recovery.
> Add an adapter of region failover NG for legacy scheduler
> ---------------------------------------------------------
>
> Key: FLINK-12876
> URL: https://issues.apache.org/jira/browse/FLINK-12876
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Coordination
> Affects Versions: 1.9.0
> Reporter: Zhu Zhu
> Assignee: Zhu Zhu
> Priority: Major
> Fix For: 1.9.0
>
>
> We need an adapter to adapt flip1.RestartPipelinedRegionStrategy for legacy
> scheduler, so that the legacy scheduler can support fine grained recovery.
> The failover recovery should respect scheduleMode(EAGER/LAZY_FROM_SOURCES) to
> avoid more tasks get re-scheduled than that before failover.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
