[
https://issues.apache.org/jira/browse/FLINK-7687?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16191491#comment-16191491
]
Elias Levy commented on FLINK-7687:
-----------------------------------
The [The JobManager
High-Availability|https://ci.apache.org/projects/flink/flink-docs-release-1.3/setup/jobmanager_high_availability.html]
page.
bq. To enable JobManager High Availability you have to set the
high-availability mode to zookeeper, configure a ZooKeeper quorum and set up a
masters file with all JobManagers hosts and their web UI ports.
The page mentions that you must configure the masters file to use the cluster
start-up scripts, but it is no clear the masters file is not used for other
purposes and is not needed if you don't use the cluster scripts.
> Clarify the master and slaves files are not necessary unless using the
> cluster start/stop scripts
> -------------------------------------------------------------------------------------------------
>
> Key: FLINK-7687
> URL: https://issues.apache.org/jira/browse/FLINK-7687
> Project: Flink
> Issue Type: Improvement
> Components: Documentation
> Affects Versions: 1.3.2
> Reporter: Elias Levy
> Priority: Minor
>
> It would be helpful if the documentation was clearer on the fact that the
> master/slaves config files are not needed when configured in
> high-availability mode unless you are using the provided scripts to start and
> shutdown the cluster over SSH. If you are using some other mechanism to
> manage Flink instances (configuration management tools such as Chef or
> Ansible, or container management frameworks like Docker Compose or
> Kubernetes), these files are unnecessary.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
