Return-Path: X-Original-To: apmail-hc-dev-archive@www.apache.org Delivered-To: apmail-hc-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 6731110A8B for ; Fri, 31 May 2013 14:03:35 +0000 (UTC) Received: (qmail 12176 invoked by uid 500); 31 May 2013 14:03:35 -0000 Delivered-To: apmail-hc-dev-archive@hc.apache.org Received: (qmail 10614 invoked by uid 500); 31 May 2013 14:03:24 -0000 Mailing-List: contact dev-help@hc.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: "HttpComponents Project" Delivered-To: mailing list dev@hc.apache.org Received: (qmail 10225 invoked by uid 99); 31 May 2013 14:03:22 -0000 Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 31 May 2013 14:03:22 +0000 Date: Fri, 31 May 2013 14:03:22 +0000 (UTC) From: "Oleg Kalnichevski (JIRA)" To: dev@hc.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (HTTPCORE-341) As a developer I want NIO Pool to close 'slow' http requests automatically 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/HTTPCORE-341?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13671481#comment-13671481 ] Oleg Kalnichevski commented on HTTPCORE-341: -------------------------------------------- The problem with this approach is that forces the connection manager to iterate over the entire queue of pending requests as well (which can be rather long) degrading performance as a result. I can apply the fix to the deprecated connection manager but would rather not have to do it. I am currently busy preparing HttpClient 4.3-beta2 release and will only be able to work on this issue once the release is out. If you want to speed things up, feel free to give me a helping hand. This whole thing is really pretty trivial. Oleg > As a developer I want NIO Pool to close 'slow' http requests automatically > -------------------------------------------------------------------------- > > Key: HTTPCORE-341 > URL: https://issues.apache.org/jira/browse/HTTPCORE-341 > Project: HttpComponents HttpCore > Issue Type: New Feature > Affects Versions: 4.2.2, 4.3-beta3 > Reporter: Ignat Alexeyenko > Priority: Critical > Labels: ha > Attachments: HTTPCORE-341_-_closing_pending_connection_before_new_lease.patch, JRADEV-21355-httpcore-nio-process-slow-requests_patch.patch > > > Currently the AbstractNIOConnPool is capable to handle timeouts for en-queued requests (requests added in leasingRequests connection). > What AbstractNIOConnPool - deal with slow connections, that already are open and being executed. Slow connections are connections that are not completed withing some request processing timeout. > Example. Consider a specific case, when a remote server holds a connection open, but transmits information slowly (e.g. 8 bytes in each 100ms). In application under the high load it means that we would stuck with no free connections to process new requests for a long period of time. > In such cases, as a developer I want to specify time in which we would close current open connection, even if it is already processing the data. -- 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 --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org For additional commands, e-mail: dev-help@hc.apache.org