From owner-freebsd-hackers@FreeBSD.ORG Thu Dec 4 12:26:26 2014 Return-Path: Delivered-To: freebsd-hackers@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 51FBC4F2 for ; Thu, 4 Dec 2014 12:26:26 +0000 (UTC) Received: from kerio.tuxis.nl (alcyone.saas.tuxis.net [31.3.111.19]) (using TLSv1.1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id DF3A7B60 for ; Thu, 4 Dec 2014 12:26:24 +0000 (UTC) X-Footer: dHV4aXMubmw= Received: from [31.3.104.222] ([31.3.104.222]) by kerio.tuxis.nl (Kerio Connect 8.4.0) for freebsd-hackers@freebsd.org; Thu, 4 Dec 2014 13:26:15 +0100 From: "Mark Schouten" Subject: Re: Slow nfsd write performance, tweaks needed To: freebsd-hackers@freebsd.org Organization: Tuxis Internet Engineering In-Reply-To: <2826701214-10966@kerio.tuxis.nl> Message-ID: <20141204122615.b85ad6cf@kerio.tuxis.nl> Date: Thu, 04 Dec 2014 13:26:15 +0100 X-Mailer: Kerio Connect 8.4.0 WebMail X-User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/38.0.2125.101 Safari/537.36 MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 04 Dec 2014 12:26:26 -0000 Hi, > The box is stable, but performs terribly. Surely, I'm doing something = wrong, > but I would like some tips and tricks to speed things up. I thought it would be nice to update you on this issue. I performed an upgrade to 10.1 on the productionbox, which helped in my = test-setup. But it didn't in my productionsetup. However, it did show an= other issue: da4p1 ONLINE 0 0 0 block size: 512B configured, 4096B= native Which I now fixed with a new pool and ashift=3D12. That fixed a lot. I c= an now write to the box, filling the gigabit with NFS traffic and not sh= owing an enormous delay in replies. I'm also quite sure that the testbox= had ashift=3D12... All other changes did not influence performance much. Now, there's only one issue left: Enabling cache-devices activates a mem= oryleak (I think)... --=C2=A0 Kerio Operator in de Cloud=3F https://www.kerioindecloud.nl/ Mark Schouten | Tuxis Internet Engineering KvK:=C2=A061527076=C2=A0| http://www.tuxis.nl/ T: 0318 200208 | info@tuxis.nl