From owner-freebsd-amd64@FreeBSD.ORG Wed Mar 9 13:43:46 2005 Return-Path: Delivered-To: freebsd-amd64@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 93BBB16A4CF for ; Wed, 9 Mar 2005 13:43:46 +0000 (GMT) Received: from buxton.digitalspy.co.uk (buxton.digitalspy.co.uk [212.42.1.208]) by mx1.FreeBSD.org (Postfix) with ESMTP id 158E843D46 for ; Wed, 9 Mar 2005 13:43:46 +0000 (GMT) (envelope-from alan_jay_uk@yahoo.co.uk) Received: from AJDELL9200 (213-78-6-149.uk.onetel.net.uk [213.78.6.149]) by buxton.digitalspy.co.uk (Postfix) with ESMTP id 017B554834 for ; Wed, 9 Mar 2005 13:43:44 +0000 (GMT) From: "Alan Jay" To: Date: Wed, 9 Mar 2005 13:43:43 -0000 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook, Build 11.0.6353 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 Thread-Index: AcUkn8AGHWxE2dSlTzel1EeKIEqGfgADcWqw In-Reply-To: <20050309120020.DBD4416A51F@hub.freebsd.org> Message-Id: <20050309134344.017B554834@buxton.digitalspy.co.uk> Subject: Re: Possible problems with Broadcom BCM5704C 10/100/1000 on TyanThunder K8S pro S2882 twin Operteron X-BeenThere: freebsd-amd64@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Porting FreeBSD to the AMD64 platform List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 09 Mar 2005 13:43:46 -0000 > On Mar 4, 2005, at 7:56 AM, Alan Jay wrote: > > > We switched off the broadcom Ethernet controller (via software) and > > routed all > > traffic via the boards Intel 10/000 Ethernet connection. One of the > > problem > > databases has now been running for a few days and so far has been ok. > > > > FWIW, I have a dual opteron tyan 2881 board with 4GB and the same NIC > chip (as identified by FreeBSD 5-STABLE). No problems under pretty > decent load. Perhaps it has to do also with > 4Gb RAM. Could you > test that theory too? [Alan Jay] Yes this is the one thing left on the test options that I can think of is to switch off some the memory and try again. Since I posted the above we have done some more tests and there still seems to be a problem but we are now heading toward the view that is more memory related.