[
https://issues.apache.org/jira/browse/FLINK-4516?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15440828#comment-15440828
]
ASF GitHub Bot commented on FLINK-4516:
---------------------------------------
GitHub user beyond1920 opened a pull request:
https://github.com/apache/flink/pull/2427
[FLINK-4516] [cluster management]leader election of resourcemanager
This pull request is to implement resourceManager leader Election, which
including:
1. When a resourceManager is started, it starts the leadership election
service first and take part in contending for leadership
2. Every resourceManager contains a ResourceManagerLeaderContender, when it
is granted leadership, it will start SlotManager and other main components.
when it is revoked leadership, it will stop all its components and clear
everything.
Main difference are 3 points:
1. Add ResourceManagerLeaderContender
2. Add getResourceManagerLeaderElectionService method in
HighAvailabilityServices, NonHaServices, TestingHighAvailabilityServices to get
leadership election service for resourceManager
3. Add a test for ResourceManager HA
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/alibaba/flink jira-4516
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/flink/pull/2427.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #2427
----
commit 3117989b87c5e9a002251353a47271fed0a84271
Author: beyond1920 <beyond1...@126.com>
Date: 2016-08-27T06:14:28Z
leader election of resourcemanager
----
> ResourceManager leadership election
> -----------------------------------
>
> Key: FLINK-4516
> URL: https://issues.apache.org/jira/browse/FLINK-4516
> Project: Flink
> Issue Type: Sub-task
> Components: Cluster Management
> Reporter: zhangjing
> Assignee: zhangjing
>
> 1. When a resourceManager is started, it starts the leadership election
> service first and take part in contending for leadership
> 2. Every resourceManager contains a ResourceManagerLeaderContender, when it
> is granted leadership, it will start SlotManager and other main components.
> when it is revoked leadership, it will stop all its components and clear
> everything.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
