From owner-freebsd-stable Fri Dec 28 5: 5:14 2001 Delivered-To: freebsd-stable@freebsd.org Received: from cpu.ath.cx (AAnnecy-101-1-4-216.abo.wanadoo.fr [193.251.17.216]) by hub.freebsd.org (Postfix) with ESMTP id D4B7E37B416 for ; Fri, 28 Dec 2001 05:05:10 -0800 (PST) Received: (from gradur@localhost) by cpu.ath.cx (8.11.6/8.11.6) id fBSD3AU02355 for freebsd-stable@freebsd.org; Fri, 28 Dec 2001 14:03:10 +0100 (CET) (envelope-from gradur) Date: Fri, 28 Dec 2001 14:03:10 +0100 From: "Julien B." To: freebsd-stable@freebsd.org Subject: NFS crashs with low kern.ipc.nmbclusters Message-ID: <20011228140310.A2232@harimandir> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG I have recently had my system crashing under high NFS load. I could reproduce it with no problem, and it seams to happen with 1 NFSv3 linux client, uploading big files (the crash happens after 300-400 Mbytes). Note that i've had the same "success" with -STABLE and 4.4-RELEASE. Just before crashing, syslog says "m_clalloc failed, consider increase NMBCLUSTERS value", and the system crashs a couple of seconds later. I did change the value of NMBCLUSTERS to 8192 (default is something around 1024), and it really fixed that problem. Is this usual to see systems crashing after unsuccessful m_clalloc() calls ? -- Regards Julien B To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message