From owner-freebsd-stable@FreeBSD.ORG Fri Nov 6 20:08:28 2009 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 7A2F9106568D for ; Fri, 6 Nov 2009 20:08:28 +0000 (UTC) (envelope-from mloftis@wgops.com) Received: from juggler.wgops.com (juggler.wgops.com [204.11.247.41]) by mx1.freebsd.org (Postfix) with ESMTP id 51FF18FC1F for ; Fri, 6 Nov 2009 20:08:28 +0000 (UTC) Received: by juggler.wgops.com (Postfix, from userid 65534) id D5504A810E; Fri, 6 Nov 2009 12:50:42 -0700 (MST) X-Spam-ASN: X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on juggler.wgops.com X-Spam-Level: X-Spam-Status: No, score=-1.3 required=5.0 tests=ALL_TRUSTED, SARE_SUB_OBFU_OTHER autolearn=no version=3.2.5 Received: from [192.168.1.44] (host-72-174-39-176.msl-mt.client.bresnan.net [72.174.39.176]) by juggler.wgops.com (Postfix) with ESMTPSA id CD10AA8070 for ; Fri, 6 Nov 2009 12:50:40 -0700 (MST) Date: Fri, 06 Nov 2009 12:50:45 -0700 From: Michael Loftis To: freebsd-stable@freebsd.org Message-ID: <8378FC0601DCAA4F0E88C577@[192.168.1.44]> In-Reply-To: References: <772532900-1257123963-cardhu_decombobulator_blackberry.rim.net-1402739480-@bda715.bisx.prod.on.blackberry> X-Mailer: Mulberry/4.0.8 (Win32) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline X-Virus-Scanned: clamav-milter 0.95.2 at juggler X-Virus-Status: Clean Subject: Re: Performance issues with 8.0 ZFS and sendfile/lighttpd X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 06 Nov 2009 20:08:28 -0000 --On Monday, November 02, 2009 10:52 AM +0100 Ivan Voras wrote: > AFAIK, ZFS is incompatible (currently) with some advanced VM operations > (like mmap, and I think sendfile relies on the same mechanism as mmap), > so that could be a cause of the slowdown. Though I'm surprised you can > only get 200 MBit/s - that's 25 MB/s and I think that even with multiple > memcpy-ing data around the kernel you should be able to get hundreds of > MB/s on newer hardware (which normally really can achieve tens of > gigabytes/s of sustained memory access). > > What else can you observe from your system? Do you have exceedingly high > sys times and load numbers? I'm also interested in what does 10 seconds > of running 'vmstat 1' looks like on your system. Is it a bare machine or > a virtual machine? Real hardware, dual quad core opteron with 64GB memory, a 3Ware 9650SE for disks. The rest of the machine is not doing much if anything when the issue happens. I've had to remove ZFS from all of the media streaming servers. I ca probably get one up for testing again over the next few weeks. I've some more hardware coming in.