Return-Path: X-Original-To: archive-asf-public-internal@cust-asf2.ponee.io Delivered-To: archive-asf-public-internal@cust-asf2.ponee.io Received: from cust-asf.ponee.io (cust-asf.ponee.io [163.172.22.183]) by cust-asf2.ponee.io (Postfix) with ESMTP id 8AB8A200AC8 for ; Mon, 23 May 2016 23:57:19 +0200 (CEST) Received: by cust-asf.ponee.io (Postfix) id 87790160A0E; Mon, 23 May 2016 21:57:19 +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 D64DB160A05 for ; Mon, 23 May 2016 23:57:18 +0200 (CEST) Received: (qmail 32169 invoked by uid 500); 23 May 2016 21:57:13 -0000 Mailing-List: contact dev-help@reef.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@reef.apache.org Delivered-To: mailing list dev@reef.apache.org Received: (qmail 31973 invoked by uid 99); 23 May 2016 21:57:13 -0000 Received: from arcas.apache.org (HELO arcas) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 23 May 2016 21:57:13 +0000 Received: from arcas.apache.org (localhost [127.0.0.1]) by arcas (Postfix) with ESMTP id D64B32C1F5C for ; Mon, 23 May 2016 21:57:12 +0000 (UTC) Date: Mon, 23 May 2016 21:57:12 +0000 (UTC) From: "Markus Weimer (JIRA)" To: dev@reef.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (REEF-1399) Node stuck in group communication failure case MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 archived-at: Mon, 23 May 2016 21:57:19 -0000 [ https://issues.apache.org/jira/browse/REEF-1399?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15297173#comment-15297173 ] Markus Weimer commented on REEF-1399: ------------------------------------- I'm with Dhruv on this. Also, I am equally ignorant to network programming. But real time units have a way of messing with us in the long run, when machines are unusually slow or fast compared to our expectations. Time-outs on the worker nodes also mess with the REEF promise of a centralized control flow. Hence, I strongly favor putting the Driver in charge of shutting down Tasks that need to be shut down. > Node stuck in group communication failure case > ---------------------------------------------- > > Key: REEF-1399 > URL: https://issues.apache.org/jira/browse/REEF-1399 > Project: REEF > Issue Type: Bug > Reporter: Julia > Labels: FT > > Currently, in the group communication, if one of the task fails, all the other tasks are waiting forever, that could easily cause leak as those tasks are running in separate threads. > There are two ways to resolve it: > 1. Add time out in the blocking call in GC. After waiting for longer enough and still not able to receive any message, throw Group Communication exception. > 2. Depend on fault tolerant to let driver to send close event to those tasks, when the task is not iterating and hung, after a timeout, enforce the task to close by throwing exception. > We will do the second in any case. Question is shall we do the first one? -- This message was sent by Atlassian JIRA (v6.3.4#6332)