From yarn-issues-return-136936-archive-asf-public=cust-asf.ponee.io@hadoop.apache.org Thu Feb 1 14:19:13 2018 Return-Path: X-Original-To: archive-asf-public@eu.ponee.io Delivered-To: archive-asf-public@eu.ponee.io Received: from cust-asf.ponee.io (cust-asf.ponee.io [163.172.22.183]) by mx-eu-01.ponee.io (Postfix) with ESMTP id 76787180652 for ; Thu, 1 Feb 2018 14:19:13 +0100 (CET) Received: by cust-asf.ponee.io (Postfix) id 6638A160C44; Thu, 1 Feb 2018 13:19:13 +0000 (UTC) Delivered-To: archive-asf-public@cust-asf.ponee.io Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by cust-asf.ponee.io (Postfix) with SMTP id 87C84160C26 for ; Thu, 1 Feb 2018 14:19:12 +0100 (CET) Received: (qmail 5309 invoked by uid 500); 1 Feb 2018 13:19:06 -0000 Mailing-List: contact yarn-issues-help@hadoop.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Delivered-To: mailing list yarn-issues@hadoop.apache.org Received: (qmail 5298 invoked by uid 99); 1 Feb 2018 13:19:06 -0000 Received: from pnap-us-west-generic-nat.apache.org (HELO spamd3-us-west.apache.org) (209.188.14.142) by apache.org (qpsmtpd/0.29) with ESMTP; Thu, 01 Feb 2018 13:19:06 +0000 Received: from localhost (localhost [127.0.0.1]) by spamd3-us-west.apache.org (ASF Mail Server at spamd3-us-west.apache.org) with ESMTP id 47EFB19A775 for ; Thu, 1 Feb 2018 13:19:06 +0000 (UTC) X-Virus-Scanned: Debian amavisd-new at spamd3-us-west.apache.org X-Spam-Flag: NO X-Spam-Score: -109.511 X-Spam-Level: X-Spam-Status: No, score=-109.511 tagged_above=-999 required=6.31 tests=[ENV_AND_HDR_SPF_MATCH=-0.5, KAM_ASCII_DIVIDERS=0.8, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_SPF_WL=-7.5, USER_IN_WHITELIST=-100] autolearn=disabled Received: from mx1-lw-us.apache.org ([10.40.0.8]) by localhost (spamd3-us-west.apache.org [10.40.0.10]) (amavisd-new, port 10024) with ESMTP id afTUuMJ4nxmR for ; Thu, 1 Feb 2018 13:19:01 +0000 (UTC) Received: from mailrelay1-us-west.apache.org (mailrelay1-us-west.apache.org [209.188.14.139]) by mx1-lw-us.apache.org (ASF Mail Server at mx1-lw-us.apache.org) with ESMTP id D01005F640 for ; Thu, 1 Feb 2018 13:19:00 +0000 (UTC) Received: from jira-lw-us.apache.org (unknown [207.244.88.139]) by mailrelay1-us-west.apache.org (ASF Mail Server at mailrelay1-us-west.apache.org) with ESMTP id 59506E024A for ; Thu, 1 Feb 2018 13:19:00 +0000 (UTC) Received: from jira-lw-us.apache.org (localhost [127.0.0.1]) by jira-lw-us.apache.org (ASF Mail Server at jira-lw-us.apache.org) with ESMTP id 187B521E81 for ; Thu, 1 Feb 2018 13:19:00 +0000 (UTC) Date: Thu, 1 Feb 2018 13:19:00 +0000 (UTC) From: "Yuqi Wang (JIRA)" To: yarn-issues@hadoop.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Updated] (YARN-7872) labeled node cannot be used to satisfy locality specified request MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/YARN-7872?page=3Dcom.atlassian= .jira.plugin.system.issuetabpanels:all-tabpanel ] Yuqi Wang updated YARN-7872: ---------------------------- Description:=20 labeled node (i.e. node with 'not empty' node label) cannot be used=C2=A0to= satisfy locality specified request (i.e. container request with 'not ANY' = resource name and the relax locality is false). For example: The node with available resource: [Resource: [MemoryMB: [100] CpuNumber: [12]] {color:#14892c}NodeLabel: [per= sistent]{color} {color:#f79232}HostName: \{SRG}{color} RackName: \{/default= -rack}] The container=C2=A0request: [Priority: [1] Resource: [MemoryMB: [1] CpuNumber: [1]] {color:#14892c}Nod= eLabel: [null]{color} {color:#f79232}HostNames: \{SRG}{color} RackNames: {}= {color:#59afe1}RelaxLocality: [false]{color}] Current RM capacity scheduler's behavior is that, the node cannot allocate = container for the request because=C2=A0of the node label not matched in the= leaf queue assign container. However, node locality and node label should be=C2=A0two orthogonal dimensi= ons to select candidate nodes for=C2=A0container=C2=A0request. And the=C2= =A0node label matching should only be executed for container request with A= NY resource name, since only this kind of container request is allowed to h= ave 'not empty'=C2=A0node label. So, for=C2=A0container request with 'not ANY' resource name (besides, it sh= ould=C2=A0not have node label), we should use=C2=A0resource name to match w= ith the node instead of node label to=C2=A0match with the node. And it shou= ld be safe, since=C2=A0the node which is not accessible for the queue will= =C2=A0not be sent in the leaf queue. Attachment is the fix according to this=C2=A0principle, please help to revi= ew. Without it, we cannot use=C2=A0locality to request container within these l= abeled nodes. If the fix is acceptable, we should also recheck whether the same issue hap= pens in trunk. was: labeled node (i.e. node with 'not empty' node label) cannot be used=C2=A0to= satisfy locality specified request (i.e. container request with 'not ANY' = resource name and the relax locality is false). For example: The node with available resource: [Resource: [MemoryMB: [100] CpuNumber: [12]] {color:#14892c}NodeLabel: [per= sistent]{color} {color:#f79232}HostName: \{SRG}{color} RackName: \{/default= -rack}] The container=C2=A0request: [Priority: [1] Resource: [MemoryMB: [1] CpuNumber: [1]] {color:#14892c}Nod= eLabel: [null]{color} {color:#f79232}HostNames: \{SRG}{color} RackNames: {}= {color:#59afe1}RelaxLocality: [false]{color}] Current RM capacity scheduler's behaiour is: The node cannot allocate container for the request because=C2=A0of the nod= e label not matched in the leaf queue assign container. However, node locality and node label should be=C2=A0two orthogonal dimensi= ons to select candidate nodes for=C2=A0container=C2=A0request. And the=C2= =A0node label matching should only be executed for container request with A= NY resource name, since only this kind of container request is allowed to h= ave 'not empty'=C2=A0node label. So, for=C2=A0container request with 'not ANY' resource name (besides, it sh= ould=C2=A0not have node label), we should use=C2=A0resource name to match w= ith the node instead of node label to=C2=A0match with the node. And it shou= ld be safe, since=C2=A0the node which is not accessible for the queue will= =C2=A0not be sent in the leaf queue. Attachment is the fix according to this=C2=A0principle, please help to revi= ew. Without it, we cannot use=C2=A0locality to request container within these l= abeled nodes. If the fix is acceptable, we should also recheck whether the same issue hap= pens in trunk. > labeled node cannot be used to satisfy locality specified request > ----------------------------------------------------------------- > > Key: YARN-7872 > URL: https://issues.apache.org/jira/browse/YARN-7872 > Project: Hadoop YARN > Issue Type: Bug > Components: capacity scheduler, capacityscheduler, resourcemanag= er > Affects Versions: 2.7.2 > Reporter: Yuqi Wang > Assignee: Yuqi Wang > Priority: Blocker > Fix For: 2.8.0, 2.7.2 > > > labeled node (i.e. node with 'not empty' node label) cannot be used=C2=A0= to satisfy locality specified request (i.e. container request with 'not ANY= ' resource name and the relax locality is false). > For example: > The node with available resource: > [Resource: [MemoryMB: [100] CpuNumber: [12]] {color:#14892c}NodeLabel: [p= ersistent]{color} {color:#f79232}HostName: \{SRG}{color} RackName: \{/defau= lt-rack}] > The container=C2=A0request: > [Priority: [1] Resource: [MemoryMB: [1] CpuNumber: [1]] {color:#14892c}N= odeLabel: [null]{color} {color:#f79232}HostNames: \{SRG}{color} RackNames: = {} {color:#59afe1}RelaxLocality: [false]{color}] > Current RM capacity scheduler's behavior is that, the node cannot allocat= e container for the request because=C2=A0of the node label not matched in t= he leaf queue assign container. > However, node locality and node label should be=C2=A0two orthogonal dimen= sions to select candidate nodes for=C2=A0container=C2=A0request. And the=C2= =A0node label matching should only be executed for container request with A= NY resource name, since only this kind of container request is allowed to h= ave 'not empty'=C2=A0node label. > So, for=C2=A0container request with 'not ANY' resource name (besides, it = should=C2=A0not have node label), we should use=C2=A0resource name to match= with the node instead of node label to=C2=A0match with the node. And it sh= ould be safe, since=C2=A0the node which is not accessible for the queue wil= l=C2=A0not be sent in the leaf queue. > Attachment is the fix according to this=C2=A0principle, please help to re= view. > Without it, we cannot use=C2=A0locality to request container within these= labeled nodes. > If the fix is acceptable, we should also recheck whether the same issue h= appens in trunk. -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org For additional commands, e-mail: yarn-issues-help@hadoop.apache.org