[
https://issues.apache.org/jira/browse/YARN-3736?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14643314#comment-14643314
]
Anubhav Dhoot commented on YARN-3736:
-------------------------------------
Thanks [~subru] for the review.
Rebased the patch add addressed all feedback except the ZKRMstateStore
For deleting reservation, in the first step we remove the reservation and then
we check if the plan state is empty because of this removal. If so we remove
the plan state as well. Because of this two steps we cannot use a transaction
for both. So we either use 2 different transactions or leave it as is. I have
left it as is for now and let me know if you feel we still need to use 2
SafeTransactions or do something else.
> Persist the Plan information, ie. accepted reservations to the RMStateStore
> for failover
> ----------------------------------------------------------------------------------------
>
> Key: YARN-3736
> URL: https://issues.apache.org/jira/browse/YARN-3736
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: capacityscheduler, fairscheduler, resourcemanager
> Reporter: Subru Krishnan
> Assignee: Anubhav Dhoot
> Attachments: YARN-3736.001.patch, YARN-3736.001.patch
>
>
> We need to persist the current state of the plan, i.e. the accepted
> ReservationAllocations & corresponding RLESpareseResourceAllocations to the
> RMStateStore so that we can recover them on RM failover. This involves making
> all the reservation system data structures protobuf friendly.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
