hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Chris Douglas (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-5525) Provide a way to query job tracker about its daemon thread's status
Date Fri, 27 Mar 2009 06:49:50 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-5525?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12689832#action_12689832
] 

Chris Douglas commented on HADOOP-5525:
---------------------------------------

This seems like it should be part of the {{ping}} functionality in HADOOP-3628 instead of
an addition to a status summary. It's more about JT health than the state of the cluster...

More to the point, if critical JT threads are dying, adding an admin tool to ask after their
state seems like the wrong response. What is this in support of?

> Provide a way to query job tracker about its daemon thread's status
> -------------------------------------------------------------------
>
>                 Key: HADOOP-5525
>                 URL: https://issues.apache.org/jira/browse/HADOOP-5525
>             Project: Hadoop Core
>          Issue Type: New Feature
>          Components: mapred
>            Reporter: Amareshwari Sriramadasu
>            Assignee: Amareshwari Sriramadasu
>             Fix For: 0.21.0
>
>         Attachments: patch-5525.txt
>
>
> Admin needs to know the status of all threads in JobTracker (whether they are alive or
not) at any point of time. This helps alot in debugging crashes.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message