[jira] [Assigned] (HIVE-20837) Tez AM container do not get killed after killing hive jvm when HS2 is not up

["mrunmayi Sahasrabudhe (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HIVE-20837?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

mrunmayi Sahasrabudhe reassigned HIVE-20837:
--------------------------------------------

    Assignee: mrunmayi Sahasrabudhe

> Tez AM container do not get killed after killing hive jvm when HS2 is not up
> ----------------------------------------------------------------------------
>
>                 Key: HIVE-20837
>                 URL: https://issues.apache.org/jira/browse/HIVE-20837
>             Project: Hive
>          Issue Type: Bug
>          Components: Hive
>    Affects Versions: 2.3.2
>            Reporter: mrunmayi Sahasrabudhe
>            Assignee: mrunmayi Sahasrabudhe
>            Priority: Major
>
> Reproduction steps :
> 1. Start hive by setting execution engine to tez
> 3. Run any command 
> 4. yarn application -list 
> 5. Kill the command before it finishes.
> 6. check in yarn application -list for the same application id. It is visible till upto 10mins and gets shutdown due to timeout.
> Ideally it should kill all the non default sessions.
> In case of Non HS2 all the session are non default.
> Other details:
> 1. This issue is reproducible with hive-2.1.1 and not in hive1.2
> 3. Does status of application should be shown as killed or succeed should be decided.
> for that refer QTEZ-137
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)