From owner-freebsd-stable Thu Oct 3 8:47: 4 2002 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 DE11937B404 for ; Thu, 3 Oct 2002 08:47:02 -0700 (PDT) Received: from spitfire.velocet.net (spitfire.velocet.net [216.138.223.227]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2946C43E6A for ; Thu, 3 Oct 2002 08:47:02 -0700 (PDT) (envelope-from steve@nomad.tor.lets.net) Received: from nomad.tor.lets.net (H74.C220.tor.velocet.net [216.138.220.74]) by spitfire.velocet.net (Postfix) with SMTP id A6852FB4576 for ; Thu, 3 Oct 2002 11:47:00 -0400 (EDT) Received: (qmail 48725 invoked by uid 1001); 3 Oct 2002 15:42:02 -0000 Date: Thu, 3 Oct 2002 11:42:02 -0400 From: Steve Shorter To: stable@freebsd.org Subject: Tuning kern.polling Message-ID: <20021003114202.A48705@nomad.lets.net> 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 Howdy! I was doing some stress testing of a SCSI chain on a 4.6.2 system with 2 big RAID volumes by just having lots of processes moving lots of data over the SCSI buss. The box has fxp network interfaces with polling enabled. the kernel config has options HZ=1000. kern.polling.user_frac = 50 During the testing the following messages started to appear in the logs. /kernel: poll stalled [65] in phase 2 /kernel: poll stalled [66] in phase 2 /kernel: poll stalled [67] in phase 2 .. etc Also I was unable to ssh to the box during this period. So I figure that the polling system needs to be tuned. Do I need increase the clock (options HZ) and/or increase/decrease kern.polling.user_frac? or is this symtpmatic of some other problem? -steve To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message