Mailing-List: contact yarn-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: yarn-issues@hadoop.apache.org
Date: Wed, 8 Apr 2015 19:00:13 +0000 (UTC)
From: "Thomas Graves (JIRA)" <jira@apache.org>
To: yarn-issues@hadoop.apache.org
Message-ID: <JIRA.12787696.1427984379000.30267.1428519613314@Atlassian.JIRA>
In-Reply-To: <JIRA.12787696.1427984379000@Atlassian.JIRA>
References: <JIRA.12787696.1427984379000@Atlassian.JIRA>
 <JIRA.12787696.1427984379849@arcas>
Subject: [jira] [Comment Edited] (YARN-3434) Interaction between
 reservations and userlimit can result in significant ULF violation
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/YARN-3434?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14485798#comment-14485798 ] 

Thomas Graves edited comment on YARN-3434 at 4/8/15 6:59 PM:
-------------------------------------------------------------

[~wangda] YARN-3243 fixes part of the problem with the max capacities, but it doesn't solve the user limit side of it.   The user limit check is never done again in assignContainer() if it skipped the checks in assignContainers() based on reservations but then is allowed to shouldAllocOrReserveNewContainer.  I'll have a patch up for this shortly I would appreciate it if you could take a look and give me feedback.


was (Author: tgraves):
[~wangda] YARN-3243 fixes part of the problem with the max capacities, but it doesn't solve the user limit side of it.   The user limit check is never done again.  I'll have a patch up for this shortly I would appreciate it if you could take a look and give me feedback.

> Interaction between reservations and userlimit can result in significant ULF violation
> --------------------------------------------------------------------------------------
>
>                 Key: YARN-3434
>                 URL: https://issues.apache.org/jira/browse/YARN-3434
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: capacityscheduler
>    Affects Versions: 2.6.0
>            Reporter: Thomas Graves
>            Assignee: Thomas Graves
>
> ULF was set to 1.0
> User was able to consume 1.4X queue capacity.
> It looks like when this application launched, it reserved about 1000 containers, each 8G each, within about 5 seconds. I think this allowed the logic in assignToUser() to allow the userlimit to be surpassed.


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