[
https://issues.apache.org/jira/browse/YARN-891?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13803437#comment-13803437
]
Jian He commented on YARN-891:
------------------------------
Had an offline discussion with Vinod and made a bunch of changes, mainly:
- create new updateApplicationStateInternal API of RMStateStore for FS/ZK state
to override for updating application state and correspondingly the update
events.
- refactor and rename some newly added methods/transitions inside RMAppImpl and
RMAppAttemptImpl
- RMAppManager.recover() is changed to always recover applications, let
RMAppImpl transition internally decide whether to launch the application or not.
- Add more unit tests in TestRMRestart for getting applications report / list
after RM restarts.
- Add test for FS/ZK state store to verify newly added fields are persisted
well.
To do:
- We should move the newInstance methods from both the data PM impls to the
data objects themselves.
- Single node test with ZK store.
> Store completed application information in RM state store
> ---------------------------------------------------------
>
> Key: YARN-891
> URL: https://issues.apache.org/jira/browse/YARN-891
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: resourcemanager
> Reporter: Bikas Saha
> Assignee: Jian He
> Attachments: YARN-891.1.patch, YARN-891.2.patch, YARN-891.3.patch,
> YARN-891.patch, YARN-891.patch, YARN-891.patch, YARN-891.patch,
> YARN-891.patch, YARN-891.patch
>
>
> Store completed application/attempt info in RMStateStore when
> application/attempt completes. This solves some problems like finished
> application get lost after RM restart and some other races like YARN-1195
--
This message was sent by Atlassian JIRA
(v6.1#6144)
