Return-Path: X-Original-To: apmail-hive-dev-archive@www.apache.org Delivered-To: apmail-hive-dev-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id A9F1F951C for ; Thu, 14 Mar 2013 00:52:17 +0000 (UTC) Received: (qmail 23349 invoked by uid 500); 14 Mar 2013 00:52:17 -0000 Delivered-To: apmail-hive-dev-archive@hive.apache.org Received: (qmail 23294 invoked by uid 500); 14 Mar 2013 00:52:17 -0000 Mailing-List: contact dev-help@hive.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@hive.apache.org Delivered-To: mailing list dev@hive.apache.org Received: (qmail 23226 invoked by uid 500); 14 Mar 2013 00:52:16 -0000 Delivered-To: apmail-hadoop-hive-dev@hadoop.apache.org Received: (qmail 23190 invoked by uid 99); 14 Mar 2013 00:52:16 -0000 Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Thu, 14 Mar 2013 00:52:16 +0000 Date: Thu, 14 Mar 2013 00:52:16 +0000 (UTC) From: "Chris Drome (JIRA)" To: hive-dev@hadoop.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (HIVE-4166) closeAllForUGI causes failure in hiveserver2 when fetching large amount of data 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/HIVE-4166?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13601912#comment-13601912 ] Chris Drome commented on HIVE-4166: ----------------------------------- [~prasadm] thanks for the comment. We are working with a patched version of 0.10, but I will keep this in mind. The original code assumed it was okay to close all FileSystem resources at the end of each call to process, with HiveServer2 this assumption no longer holds. So it should be fixed to ensure that doAs works properly or doAs should be deprecated in favor of impersonation. > closeAllForUGI causes failure in hiveserver2 when fetching large amount of data > ------------------------------------------------------------------------------- > > Key: HIVE-4166 > URL: https://issues.apache.org/jira/browse/HIVE-4166 > Project: Hive > Issue Type: Bug > Components: Security, Shims > Affects Versions: 0.10.0, 0.11.0 > Reporter: Chris Drome > Assignee: Chris Drome > Attachments: HIVE-4166-0.10.patch, HIVE-4166-trunk.patch > > > HiveServer2 configured to use Kerberos authentication with doAs enabled throws an exception when fetching a large amount of data from a query. > The exception is caused because FileSystem.closeAllForUGI is always called at the end of TUGIAssumingProcessor.process. This affects requests on the ResultSet for data from a SELECT query when the amount of data exceeds a certain size. At that point any subsequent calls to fetch more data throw an exception because the underlying DFSClient has been closed. -- 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