You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@hive.apache.org by "Siddharth Seth (JIRA)" <ji...@apache.org> on 2015/04/07 22:56:13 UTC

[jira] [Updated] (HIVE-10025) LLAP: Queued work times out

     [ https://issues.apache.org/jira/browse/HIVE-10025?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Siddharth Seth updated HIVE-10025:
----------------------------------
    Attachment: HIVE-10025.1.txt

Patch to send occasional heartbeats to AMs for which there are queued tasks.
Also separates out the protocol used for communicating with AMs.

> LLAP: Queued work times out
> ---------------------------
>
>                 Key: HIVE-10025
>                 URL: https://issues.apache.org/jira/browse/HIVE-10025
>             Project: Hive
>          Issue Type: Sub-task
>            Reporter: Siddharth Seth
>            Assignee: Siddharth Seth
>             Fix For: llap
>
>         Attachments: HIVE-10025.1.txt
>
>
> If a daemon holds a task in queue for a long time, it'll eventually time out - but isn't removed from the queue. Ideally, it shouldn't be allowed to time out. Otherwise, handle the timeout so that the task doesn't run - or starts and fails - likely a change in the TaskCommunicator.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)