[
https://issues.apache.org/jira/browse/FLINK-15813?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17028690#comment-17028690
]
Zhu Zhu commented on FLINK-15813:
---------------------------------
I think we can do this after we have done step #3 in FLINK-15626(remove legacy
scheduler). In this way we will not need to fix cases for the legacy region
failover strategy and possibly fix them again when we switched the cases to run
with the DefaultScheduler.
{{DefaultScheduler}} now already uses "region" as the default failover strategy
(see {{FailoverStrategyFactoryLoader#loadFailoverStrategyFactory}}). So
FLINK-15626 step #3 will ensure that no case breaks with region failover
strategy, and in this ticket we can simply change the config option default
value to "region" and break nothing.
> Set default value of jobmanager.execution.failover-strategy to region
> ---------------------------------------------------------------------
>
> Key: FLINK-15813
> URL: https://issues.apache.org/jira/browse/FLINK-15813
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Coordination
> Affects Versions: 1.10.0
> Reporter: Till Rohrmann
> Priority: Blocker
> Fix For: 1.11.0
>
>
> We should set the default value of {{jobmanager.execution.failover-strategy}}
> to {{region}}. This might require to adapt existing tests to make them pass.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
