drill-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jinfeng Ni (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (DRILL-1592) Detect drillbit failure and cancel running queries.
Date Tue, 11 Nov 2014 01:25:33 GMT

     [ https://issues.apache.org/jira/browse/DRILL-1592?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Jinfeng Ni updated DRILL-1592:
    Attachment: 0002-DRILL-1592-Detect-drillbit-failure-and-cancel-the-af.patch

> Detect drillbit failure and cancel running queries. 
> ----------------------------------------------------
>                 Key: DRILL-1592
>                 URL: https://issues.apache.org/jira/browse/DRILL-1592
>             Project: Apache Drill
>          Issue Type: Bug
>            Reporter: Jinfeng Ni
>            Assignee: Jinfeng Ni
>         Attachments: 0002-DRILL-1592-Detect-drillbit-failure-and-cancel-the-af.patch
> When Drillbit dies or was killed,  the running queries might be hang there, and the sqlline
would also hang there. There are two cases:
> 1. The Drillbit having the foreman dies, or the foreman thread dies, but the rest of
non-root fragments are still running, even though the connection between the client and the
foreman has been lost.
> 2. The Drillbit running the non-root fragments dies, but foreman and other non-root fragments
are still running.
> In either cases, Drill should have a way to detect such node failure, and cancel the
running queries, and return to a state where client could submit a new query, or re-connect
(if connection is lost in case foreman dies) to a different foreman.

This message was sent by Atlassian JIRA

View raw message