hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Vinod Kumar Vavilapalli (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (MAPREDUCE-5110) Long task launch delays can lead to multiple parallel attempts of the task
Date Tue, 16 Apr 2013 03:50:18 GMT

    [ https://issues.apache.org/jira/browse/MAPREDUCE-5110?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13632543#comment-13632543
] 

Vinod Kumar Vavilapalli commented on MAPREDUCE-5110:
----------------------------------------------------

Trying to understand this, mostly agree with what Arun said. To summarize:
 - Strictly guaranteeing serial execution of task attempts is not possible in general and
is a non-requirement
 - JT already deals with all kinds of slow-ness with tasks and irrespective of this patch,
clients have to deal with the slowness.

bq. Where possible (i.e., not transient network partitions), run a single task attempt for
a task when speculation is turned off
Seems an arbitrary non-requirement, don't see what we gain from this.

The JIRA started with the above goal which isn't worth pursing from what I see, but now it
seems to have transformed into something more benign. Looked at the patch. It looks like you
want quicker failure when tasks are getting launched/localized to meet some kind of SLAs?
If that is the case, instead of calling it a 'TT-side implementation', if we call it an aggressive
timeout enforced on TTs for tasks, and make it job-configurable, that should do. Right?
                
> Long task launch delays can lead to multiple parallel attempts of the task
> --------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-5110
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-5110
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: tasktracker
>    Affects Versions: 1.1.2
>            Reporter: Karthik Kambatla
>            Assignee: Karthik Kambatla
>         Attachments: expose-mr-5110.patch, mr-5110.patch, mr-5110.patch, mr-5110-tt-only.patch
>
>
> If a task takes too long to launch, the JT expires the task and schedules another attempt.
The earlier attempt can start after the later attempt leading to two parallel attempts running
at the same time. This is particularly an issue if the user turns off speculation and expects
a single attempt of a task to run at any point in time.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message