Return-Path: X-Original-To: apmail-aurora-issues-archive@minotaur.apache.org Delivered-To: apmail-aurora-issues-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 6F57A18757 for ; Thu, 1 Oct 2015 20:26:27 +0000 (UTC) Received: (qmail 91194 invoked by uid 500); 1 Oct 2015 20:26:27 -0000 Delivered-To: apmail-aurora-issues-archive@aurora.apache.org Received: (qmail 91135 invoked by uid 500); 1 Oct 2015 20:26:27 -0000 Mailing-List: contact issues-help@aurora.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@aurora.apache.org Delivered-To: mailing list issues@aurora.apache.org Received: (qmail 91123 invoked by uid 99); 1 Oct 2015 20:26:27 -0000 Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Thu, 01 Oct 2015 20:26:27 +0000 Date: Thu, 1 Oct 2015 20:26:27 +0000 (UTC) From: "Kevin Sweeney (JIRA)" To: issues@aurora.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (AURORA-1511) PreemptorService failure does not trigger shutdown MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/AURORA-1511?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14940363#comment-14940363 ] Kevin Sweeney commented on AURORA-1511: --------------------------------------- It looks like we don't register a Listener on our ServiceManagers to shutdown the JVM / execute a shutdown action as described in http://docs.guava-libraries.googlecode.com/git/javadoc/com/google/common/util/concurrent/ServiceManager.html. I think we should adopt our use of the library to more-or-less copy the pattern presented in ServiceManager's javadoc. I think this is a regression, as such I would block 0.10.0 until this is released. > PreemptorService failure does not trigger shutdown > -------------------------------------------------- > > Key: AURORA-1511 > URL: https://issues.apache.org/jira/browse/AURORA-1511 > Project: Aurora > Issue Type: Bug > Reporter: Zameer Manji > > While observing AURORA-1510 in production I noticed the bug caused the {{PreemptorService}} to transition to the FAILED state. The {{/services}} endpoint had: > {noformat} > { > name: "PreemptorService", > state: "FAILED", > failureCause: "java.util.ConcurrentModificationException" > }, > {noformat} > However the scheduler continued to run. I believe this is a bug. -- This message was sent by Atlassian JIRA (v6.3.4#6332)