From owner-freebsd-current@FreeBSD.ORG Thu Sep 13 22:07:45 2007 Return-Path: Delivered-To: current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id EA75716A420 for ; Thu, 13 Sep 2007 22:07:45 +0000 (UTC) (envelope-from cb@severious.net) Received: from ion.gank.org (ion.gank.org [69.55.238.164]) by mx1.freebsd.org (Postfix) with ESMTP id AD04513C45B for ; Thu, 13 Sep 2007 22:07:45 +0000 (UTC) (envelope-from cb@severious.net) Received: by ion.gank.org (Postfix, from userid 1001) id 4D39E118C9; Thu, 13 Sep 2007 17:07:45 -0500 (CDT) Date: Thu, 13 Sep 2007 17:07:44 -0500 From: Craig Boston To: Pawel Jakub Dawidek Message-ID: <20070913220744.GB5814@nowhere> Mail-Followup-To: Craig Boston , Pawel Jakub Dawidek , current@freebsd.org References: <20070906184853.GB90021@roadrunner.spoerlein.net> <20070910195833.GE1103@garage.freebsd.pl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070910195833.GE1103@garage.freebsd.pl> User-Agent: Mutt/1.4.2.3i Cc: current@freebsd.org Subject: Re: ZFS not caching right? X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 13 Sep 2007 22:07:46 -0000 On Mon, Sep 10, 2007 at 09:58:33PM +0200, Pawel Jakub Dawidek wrote: > Can you try recent HEAD? I committed a change that allows ZFS cache > (ARC) to use more memory. This should be safe after dfr@'s vnode leak > fix and should also help performance. FYI, I've started getting kmem_map too small panics again after this change, on i386. 2GB physical RAM vm.kmem_size: 671088640 (set in loader.conf) kern.maxvnodes: 75000 (down from the autotuned value of 100000) vfs.zfs.arc_min: 20971520 vfs.zfs.arc_max: 503316480 vfs.zfs.vdev.cache.size: 10485760 vfs.zfs.vdev.cache.max: 16384 Current values from a live system: kstat.zfs.misc.arcstats.c_min: 20971520 kstat.zfs.misc.arcstats.c_max: 503316480 kstat.zfs.misc.arcstats.size: 321891840 The panic showed "kmem_malloc(1568768): kmem_map too small: 458543104 total allocated" last time it happened. 640MB for kmem is pretty close to the max I can bump it up to before the system no longer boots for whatever reason. Let me know if I'm missing any relevant stats... Craig