From owner-freebsd-stable@FreeBSD.ORG Fri Dec 26 07:47:40 2003 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8542516A4CE; Fri, 26 Dec 2003 07:47:40 -0800 (PST) Received: from tea.blinkenlights.nl (tea.blinkenlights.nl [62.58.162.229]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2E8DC43D3F; Fri, 26 Dec 2003 07:47:39 -0800 (PST) (envelope-from sten@blinkenlights.nl) Received: by tea.blinkenlights.nl (Postfix, from userid 101) id 9B95A352; Fri, 26 Dec 2003 16:44:07 +0100 (CET) Received: from localhost (localhost [127.0.0.1]) by tea.blinkenlights.nl (Postfix) with ESMTP id 8201D1CA; Fri, 26 Dec 2003 16:44:07 +0100 (CET) Date: Fri, 26 Dec 2003 16:44:07 +0100 (CET) From: Sten To: Dmitry Pryanishnikov In-Reply-To: <20031225214950.M86827@atlantis.atlantis.dp.ua> Message-ID: References: <20031225214950.M86827@atlantis.atlantis.dp.ua> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: stable@freebsd.org cc: current@freebsd.org Subject: Re: [SOLVED] RE: [Backtrace] 4.9 and 5.1-RELEASE occasionly panic X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.1 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, 26 Dec 2003 15:47:40 -0000 On Thu, 25 Dec 2003, Dmitry Pryanishnikov wrote: > > Hello! > > > From: Sten > > > > I'd really love getting rid of most hardcoded memory allocations > > in the fbsd kernel, with the main bugger being NMBCLUSTERS, > > which one always forgets about untill it's too late :(. > > In solaris auto-tuning has been a top priority for a long while, > > and propper dynamic tuning should get it right ( or even better ) > > in 99% of the cases. People with other needs should be able > > to build a custom kernel. > > You need _not_ to build custom FreeBSD kernel in order to increase > NMBCLUSTERS. Both CURRENT and STABLE allows setting it from loader.conf: > > kern.ipc.nmbufs="32768" # Number of mbufs > kern.ipc.nmbclusters="16384" # Number of mbuf clusters > > But yes, you have to reboot your system if you want to raise them. true, but my point is that these kind of static limits are kind of silly. I have a machine with plenty of memory, when I start doing serious network traffic it dies, ecause freebsd doesn't use available memory. Static allocations are suboptimal and in corner cases they lead to loss of network connectivity or crashes. I do know that fixing these problems is a lot of work, which is why I suggested it as a goal for 6.0. to throw in a fully innapropriate quote : In other words, it's ok for a developer to say: "Ok, what happens if you do the above echo?" to figure out whether maybe the problem is due to excessive spurgel-production in the VM frobnicator, but it is NOT OK to say "you can tweak it for your use, so don't complain about the bad behaviour" -- Sten Spans "There is a crack in everything, that's how the light gets in." Leonard Cohen - Anthem