From owner-freebsd-current@FreeBSD.ORG Thu Nov 4 10:32:01 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5BCB016A4CE for ; Thu, 4 Nov 2004 10:32:01 +0000 (GMT) Received: from pophost.wldelft.nl (sunray.wldelft.nl [145.9.132.100]) by mx1.FreeBSD.org (Postfix) with ESMTP id A24BB43D4C for ; Thu, 4 Nov 2004 10:32:00 +0000 (GMT) (envelope-from leroy.vanlogchem@wldelft.nl) Received: (from root@localhost) by pophost.wldelft.nl (8.9.3/8.9.3vc) id LAA20362 for freebsd-current@freebsd.org; Thu, 4 Nov 2004 11:31:59 +0100 (MET) Received: from wldelft.nl (beasty [145.9.150.200]) by pophost.wldelft.nl (8.9.3/8.9.3) with ESMTP id LAA20130 for ; Thu, 4 Nov 2004 11:31:56 +0100 (MET) Message-ID: <418A0532.7020902@wldelft.nl> Date: Thu, 04 Nov 2004 11:32:18 +0100 From: Leroy van Logchem User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.6) Gecko/20040429 X-Accept-Language: en-us, en MIME-Version: 1.0 To: freebsd-current@freebsd.org Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: cvs commit: src/sys/sys buf.h X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 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, 04 Nov 2004 10:32:01 -0000 Hi Poul, These questions might not be related but I'll ask anyway. (I'am not very technical nor expect on kernel design so bare with me please) sysctl reports: -------------- kern.ipc.nmbclusters: 3072 BIO buffer 214 428K 668K 101107 2048 sysctl -a | grep -C3 "mbuf" | more Side question: Is the a case convention for the info in sysctl -a? sysctl -a | grep -C3 -i "mbuf" | more MbufClust: 2048, 3072, 256, 6, 256 Mbuf: 256, 0, 257, 133, 3562331 Q: Which of these will be dynamic by default in RELENG_5_3? Using 5.2.1 we had to tune several kernel params using trail'n'error during production time. Q: Will nmbclusters, kmem_size and kmem_scale avoid service-outages in the future? (bufferspace unavailable, out of kmem etc.) Cheers, Leroy