From owner-freebsd-smp@FreeBSD.ORG Sun May 27 19:26:40 2007 Return-Path: X-Original-To: freebsd-smp@freebsd.org Delivered-To: freebsd-smp@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id A91A416A421; Sun, 27 May 2007 19:26:40 +0000 (UTC) (envelope-from tundra@tundraware.com) Received: from eskimo.tundraware.com (eskimo.tundraware.com [66.92.130.161]) by mx1.freebsd.org (Postfix) with ESMTP id 49C6913C44C; Sun, 27 May 2007 19:26:40 +0000 (UTC) (envelope-from tundra@tundraware.com) Received: from [192.168.0.2] (viper.tundraware.com [192.168.0.2]) (authenticated bits=0) by eskimo.tundraware.com (8.13.8/8.13.8) with ESMTP id l4RJQW9r024620 (version=TLSv1/SSLv3 cipher=DHE-DSS-AES256-SHA bits=256 verify=NO); Sun, 27 May 2007 14:26:32 -0500 (CDT) (envelope-from tundra@tundraware.com) Message-ID: <4659DB64.2010608@tundraware.com> Date: Sun, 27 May 2007 14:26:28 -0500 From: Tim Daneliuk Organization: TundraWare Inc. User-Agent: Thunderbird 1.5.0.10 (Windows/20070221) MIME-Version: 1.0 To: freebsd-smp@freebsd.org, freebsd-hardware@freebsd.org Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-tundraware.com-MailScanner-Information: Please contact the ISP for more information X-tundraware.com-MailScanner: Found to be clean X-tundraware.com-MailScanner-SpamCheck: not spam, SpamAssassin (not cached, score=-4.399, required 1, autolearn=not spam, ALL_TRUSTED -1.80, BAYES_00 -2.60) X-tundraware.com-MailScanner-From: tundra@tundraware.com X-Spam-Status: No Cc: Subject: Problems w/Intel D946GZIS + Pentium D (Possible SMP Problem) X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: tundra@tundraware.com List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 27 May 2007 19:26:40 -0000 Just bought: Intel D946GZIS MOBO w/ICH7 SATA-II support WDC WD2500JS-00NCB1 250G Hard Drive that claims to do 3G/s Pentium D @ 3.2GHz In addition to the on-board peripherals, the two available PCI slots have an Adaptec 2930CU SCSI controller and a 3Com 3C901 card in them respectively. I've run into a problem that makes no sense to me and I suspect is related to how SMP works (or, then again, maybe not): 1) When booting FBSD 6.2, right after the second CPU is started the system hangs for about 45 seconds and then proceeds normally. This happens with GENERIC-Release, SMP-Stable, and SMP-Release kernels. During this wait, the floppy goes active for some reason. Other strange anomalies I've noted: 2) If I try to boot the 4.11-Release CD, it gets partway through the boot and then spits out: RTC BIOS diagnostic error 80 I've replaced the RTC battery and still get the same error. I get no such error with 6.2 3) Both the MOBO and drive are SATA-300 rated, but 6.2 insists that the drive is running at SATA-150. I have verified that the drive has no jumper forcing it into this mode. Anyone have ideas on what is going on here? (I've DAGS and found nothing like this in the archives.) -- ---------------------------------------------------------------------------- Tim Daneliuk tundra@tundraware.com PGP Key: http://www.tundraware.com/PGP/ From owner-freebsd-smp@FreeBSD.ORG Sun May 27 19:28:25 2007 Return-Path: X-Original-To: freebsd-smp@freebsd.org Delivered-To: freebsd-smp@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id D468916A41F; Sun, 27 May 2007 19:28:25 +0000 (UTC) (envelope-from tundra@tundraware.com) Received: from eskimo.tundraware.com (eskimo.tundraware.com [66.92.130.161]) by mx1.freebsd.org (Postfix) with ESMTP id 8AD3313C483; Sun, 27 May 2007 19:28:25 +0000 (UTC) (envelope-from tundra@tundraware.com) Received: from [192.168.0.2] (viper.tundraware.com [192.168.0.2]) (authenticated bits=0) by eskimo.tundraware.com (8.13.8/8.13.8) with ESMTP id l4RJ713A024411 (version=TLSv1/SSLv3 cipher=DHE-DSS-AES256-SHA bits=256 verify=NO); Sun, 27 May 2007 14:07:02 -0500 (CDT) (envelope-from tundra@tundraware.com) Message-ID: <4659D6D1.1090909@tundraware.com> Date: Sun, 27 May 2007 14:06:57 -0500 From: Tim Daneliuk Organization: TundraWare Inc. User-Agent: Thunderbird 1.5.0.10 (Windows/20070221) MIME-Version: 1.0 To: freebsd-smp@freebsd.org, freebsd-hardware@freebsd.org Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-tundraware.com-MailScanner-Information: Please contact the ISP for more information X-tundraware.com-MailScanner: Found to be clean X-tundraware.com-MailScanner-SpamCheck: not spam, SpamAssassin (not cached, score=-4.399, required 1, autolearn=not spam, ALL_TRUSTED -1.80, BAYES_00 -2.60) X-tundraware.com-MailScanner-From: tundra@tundraware.com X-Spam-Status: No Cc: Subject: Problems w/Intel D946GZIS + Pentium D (Possible SMP Problem) X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: tundra@tundraware.com List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 27 May 2007 19:28:25 -0000 Just bought: Intel D946GZIS MOBO w/ICH7 SATA-II support WDC WD2500JS-00NCB1 250G Hard Drive that claims to do 3G/s Pentium D @ 3.2GHz In addition to the on-board peripherals, the two available PCI slots have an Adaptec 2930CU SCSI controller and a 3Com 3C901 card in them respectively. I've run into a problem that makes no sense to me and I suspect is related to how SMP works (or, then again, maybe not): 1) When booting FBSD 6.2, right after the second CPU is started the system hangs for about 45 seconds and then proceeds normally. This happens with GENERIC-Release, SMP-Stable, and SMP-Release kernels. Other strange anomalies I've noted: 2) If I try to boot the 4.11-Release CD, it gets partway through the boot and then spits out: RTC BIOS diagnostic error 80 I've replaced the RTC battery and still get the same error. I get no such error with 6.2 3) Both the MOBO and drive are SATA-300 rated, but 6.2 insists that the drive is running at SATA-150. I have verified that the drive has no jumper forcing it into this mode. Anyone have ideas on what is going on here? (I've DAGS and found nothing like this in the archives.) -- ---------------------------------------------------------------------------- Tim Daneliuk tundra@tundraware.com PGP Key: http://www.tundraware.com/PGP/ From owner-freebsd-smp@FreeBSD.ORG Mon May 28 19:37:27 2007 Return-Path: X-Original-To: freebsd-smp@freebsd.org Delivered-To: freebsd-smp@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 9D99516A4EF for ; Mon, 28 May 2007 19:37:27 +0000 (UTC) (envelope-from bsdlist@cogeco.ca) Received: from fep7.cogeco.net (smtp2.cogeco.ca [216.221.81.29]) by mx1.freebsd.org (Postfix) with ESMTP id 74B0B13C480 for ; Mon, 28 May 2007 19:37:27 +0000 (UTC) (envelope-from bsdlist@cogeco.ca) Received: from elehost-can.cogeco.ca (d141-2-106.home.cgocable.net [24.141.2.106]) by fep7.cogeco.net (Postfix) with ESMTP id 2B0FE1242; Mon, 28 May 2007 15:37:25 -0400 (EDT) X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Mon, 28 May 2007 15:39:09 -0400 To: freebsd-smp@freebsd.org From: Paul In-Reply-To: <3aaaa3a0705271620g1fc39b84ua539c8daf194c065@mail.gmail.com > References: <00b301c785b9$0bc9b180$b6db87d4@multiplay.co.uk> <20070423154745.GA872@xor.obsecurity.org> <20070423155500.413CB3A5@fep9.cogeco.net> <20070423155922.GA1156@xor.obsecurity.org> <20070424010422.9E9376A4@fep3.cogeco.net> <20070424021553.GA4224@xor.obsecurity.org> <20070424025031.F0841AE5@fep1.cogeco.net> <20070424035249.GA5598@xor.obsecurity.org> <3aaaa3a0705050423x72b78952wfd530a492c9cea14@mail.gmail.com> <20070523145728.F040562B@fep2.cogeco.net> <3aaaa3a0705271620g1fc39b84ua539c8daf194c065@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Message-Id: <20070528193726.2B0FE1242@fep7.cogeco.net> Cc: Chris Subject: Re: System Cpu Between 50-70% and need to find out why X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 28 May 2007 19:37:27 -0000 > >I will run tests on a local box but I cant be toggling the kernels on >my production servers. > >Thanks > >Chris Hi Chris, So have you removed the QUOTA feature from your production boxes? I am curious what you did here to get things working on the stable 6.2? I am amazed more people are not having an issue with this. Thanks Paul From owner-freebsd-smp@FreeBSD.ORG Mon May 28 20:45:27 2007 Return-Path: X-Original-To: freebsd-smp@freebsd.org Delivered-To: freebsd-smp@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 7B30816A47B for ; Mon, 28 May 2007 20:45:27 +0000 (UTC) (envelope-from chrcoluk@gmail.com) Received: from ug-out-1314.google.com (ug-out-1314.google.com [66.249.92.173]) by mx1.freebsd.org (Postfix) with ESMTP id 6514C13C455 for ; Mon, 28 May 2007 20:45:26 +0000 (UTC) (envelope-from chrcoluk@gmail.com) Received: by ug-out-1314.google.com with SMTP id u2so980216uge for ; Mon, 28 May 2007 13:45:25 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=Q3blIWkcNFm318N4dWP1rHlH90RNxyyHmT1EFNo/QAUzoTjuChmpgW/qd3mA/+AR/GSIDi8FMtyq/q5UW6Xw5mZ0A0dmOkRMmLkDuoyXIBXRHTqLfSUEboVB6saZJEAWKjIizUwcn0z8iZuBewEbVjUp52Yfi90n2MR+RMTBZB4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=UdYkT0GDQm45/1D3aqzTqcAKlIhkTOHfKMMxBVUlZKniEFGEEks/RdGlSuZd2iQTvMiQ+oyWZvKtFmTYs8IIPFhhoUoX3ZKVsxg4l+F2uqo6uzGhTZZkX+ldxyXVI0S6K56lYhZbk2lK92vrzI5sxh+GptJvPXfbRugPqPNryTc= Received: by 10.66.217.5 with SMTP id p5mr2850664ugg.1180385125413; Mon, 28 May 2007 13:45:25 -0700 (PDT) Received: by 10.67.24.16 with HTTP; Mon, 28 May 2007 13:45:25 -0700 (PDT) Message-ID: <3aaaa3a0705281345k7e3244e8v75e42f3d55884b00@mail.gmail.com> Date: Mon, 28 May 2007 21:45:25 +0100 From: Chris To: Paul In-Reply-To: <20070528193726.2B0FE1242@fep7.cogeco.net> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <00b301c785b9$0bc9b180$b6db87d4@multiplay.co.uk> <20070423155922.GA1156@xor.obsecurity.org> <20070424010422.9E9376A4@fep3.cogeco.net> <20070424021553.GA4224@xor.obsecurity.org> <20070424025031.F0841AE5@fep1.cogeco.net> <20070424035249.GA5598@xor.obsecurity.org> <3aaaa3a0705050423x72b78952wfd530a492c9cea14@mail.gmail.com> <20070523145728.F040562B@fep2.cogeco.net> <3aaaa3a0705271620g1fc39b84ua539c8daf194c065@mail.gmail.com> <20070528193726.2B0FE1242@fep7.cogeco.net> Cc: freebsd-smp@freebsd.org Subject: Re: System Cpu Between 50-70% and need to find out why X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 28 May 2007 20:45:27 -0000 On 28/05/07, Paul wrote: > > > > >I will run tests on a local box but I cant be toggling the kernels on > >my production servers. > > > >Thanks > > > >Chris > > Hi Chris, > > So have you removed the QUOTA feature from your production boxes? I > am curious what you did here to get things working on the stable 6.2? > > I am amazed more people are not having an issue with this. > > Thanks > > Paul > > no I use QUOTA but put up with the performance hit, all my production servers are dual core now and new ones will be core 2 duos so I work round it with raw power, also using SATA hds instead of PATA. I dont use QUOTA on every single server tho, just the commercial web servers. Chris From owner-freebsd-smp@FreeBSD.ORG Tue May 29 11:45:28 2007 Return-Path: X-Original-To: freebsd-smp@freebsd.org Delivered-To: freebsd-smp@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id A469B16A41F for ; Tue, 29 May 2007 11:45:28 +0000 (UTC) (envelope-from bsdlist@cogeco.ca) Received: from fep7.cogeco.net (smtp2.cogeco.ca [216.221.81.29]) by mx1.freebsd.org (Postfix) with ESMTP id 7BDD913C447 for ; Tue, 29 May 2007 11:45:28 +0000 (UTC) (envelope-from bsdlist@cogeco.ca) Received: from elehost-can.cogeco.ca (d141-2-106.home.cgocable.net [24.141.2.106]) by fep7.cogeco.net (Postfix) with ESMTP id 8B7421B8F; Tue, 29 May 2007 07:45:27 -0400 (EDT) X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Tue, 29 May 2007 07:47:12 -0400 To: Chris From: Paul In-Reply-To: <3aaaa3a0705281345k7e3244e8v75e42f3d55884b00@mail.gmail.com > References: <00b301c785b9$0bc9b180$b6db87d4@multiplay.co.uk> <20070423155922.GA1156@xor.obsecurity.org> <20070424010422.9E9376A4@fep3.cogeco.net> <20070424021553.GA4224@xor.obsecurity.org> <20070424025031.F0841AE5@fep1.cogeco.net> <20070424035249.GA5598@xor.obsecurity.org> <3aaaa3a0705050423x72b78952wfd530a492c9cea14@mail.gmail.com> <20070523145728.F040562B@fep2.cogeco.net> <3aaaa3a0705271620g1fc39b84ua539c8daf194c065@mail.gmail.com> <20070528193726.2B0FE1242@fep7.cogeco.net> <3aaaa3a0705281345k7e3244e8v75e42f3d55884b00@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Message-Id: <20070529114527.8B7421B8F@fep7.cogeco.net> Cc: freebsd-smp@freebsd.org Subject: Re: System Cpu Between 50-70% and need to find out why X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 29 May 2007 11:45:28 -0000 At 04:45 PM 28/05/2007, Chris wrote: >On 28/05/07, Paul wrote: >> >> > >> >I will run tests on a local box but I cant be toggling the kernels on >> >my production servers. >> > >> >Thanks >> > >> >Chris >> >>Hi Chris, >> >>So have you removed the QUOTA feature from your production boxes? I >>am curious what you did here to get things working on the stable 6.2? >> >>I am amazed more people are not having an issue with this. >> >>Thanks >> >>Paul >> >no I use QUOTA but put up with the performance hit, all my production >servers are dual core now and new ones will be core 2 duos so I work >round it with raw power, also using SATA hds instead of PATA. I dont >use QUOTA on every single server tho, just the commercial web servers. > >Chris Thanks for your reply. That is odd as I used two Dual Core CPUS (4 in total) XEON 64 with 16 GIG of RAM and SATA HD on the amd64 branch with a very fast Areca controller and the hit was so bad that the system was not usable at all. It was like going back to a 386 (or beyond that). Did you provide any mail services on that server? I have a hunch it was the multiple instant access to many files that was causing the slowdown (50-100 of small file reads and writes per second). If you only provided FTP type access then that may not have been as big of an issue as you may not have had as many concurrent file writes. Thanks Paul From owner-freebsd-smp@FreeBSD.ORG Sat Jun 2 07:51:04 2007 Return-Path: X-Original-To: freebsd-smp@freebsd.org Delivered-To: freebsd-smp@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 008BA16A421 for ; Sat, 2 Jun 2007 07:51:04 +0000 (UTC) (envelope-from chrcoluk@gmail.com) Received: from ug-out-1314.google.com (ug-out-1314.google.com [66.249.92.171]) by mx1.freebsd.org (Postfix) with ESMTP id 64D3E13C448 for ; Sat, 2 Jun 2007 07:51:03 +0000 (UTC) (envelope-from chrcoluk@gmail.com) Received: by ug-out-1314.google.com with SMTP id u2so416634uge for ; Sat, 02 Jun 2007 00:51:02 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=tRi0N75zfb9avZtMNGgkwYM0elMlCtQ9plt+i3qhxWdqTZZErrNEZs/OlTMSb3lGqEsd4ssNP4oPXgV737xlbdw8aolZdc5kHyMvyxP1mBfyW9pvtFteJXQOopmeZAzfveHxdNDViK3v4TspqTLEJ3WesomoSxdvoW3lM8peT90= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=ez7OcUSHoQWMO4DuSwvx3VlMF2Wi/6fQxuEQ3cL23W3O6MVWH3kEHoFEO0kniL+7kEOonYOrQqzlHmzrz+CDw4YXA4gIjQmQdF20gH/1CXHzgJGpZOq6qOEOqSlFru3vaVP5R3IshS9EtDEB36lUL+wGGSvKsf38aekTDdZ3QZI= Received: by 10.67.115.19 with SMTP id s19mr1382279ugm.1180770662271; Sat, 02 Jun 2007 00:51:02 -0700 (PDT) Received: by 10.67.24.16 with HTTP; Sat, 2 Jun 2007 00:51:02 -0700 (PDT) Message-ID: <3aaaa3a0706020051v4df04911gf1a91ee59fbeb2e2@mail.gmail.com> Date: Sat, 2 Jun 2007 08:51:02 +0100 From: Chris To: Paul In-Reply-To: <20070529114527.8B7421B8F@fep7.cogeco.net> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <00b301c785b9$0bc9b180$b6db87d4@multiplay.co.uk> <20070424021553.GA4224@xor.obsecurity.org> <20070424025031.F0841AE5@fep1.cogeco.net> <20070424035249.GA5598@xor.obsecurity.org> <3aaaa3a0705050423x72b78952wfd530a492c9cea14@mail.gmail.com> <20070523145728.F040562B@fep2.cogeco.net> <3aaaa3a0705271620g1fc39b84ua539c8daf194c065@mail.gmail.com> <20070528193726.2B0FE1242@fep7.cogeco.net> <3aaaa3a0705281345k7e3244e8v75e42f3d55884b00@mail.gmail.com> <20070529114527.8B7421B8F@fep7.cogeco.net> Cc: freebsd-smp@freebsd.org Subject: Re: System Cpu Between 50-70% and need to find out why X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 02 Jun 2007 07:51:04 -0000 On 29/05/07, Paul wrote: > At 04:45 PM 28/05/2007, Chris wrote: > >On 28/05/07, Paul wrote: > >> > >> > > >> >I will run tests on a local box but I cant be toggling the kernels on > >> >my production servers. > >> > > >> >Thanks > >> > > >> >Chris > >> > >>Hi Chris, > >> > >>So have you removed the QUOTA feature from your production boxes? I > >>am curious what you did here to get things working on the stable 6.2? > >> > >>I am amazed more people are not having an issue with this. > >> > >>Thanks > >> > >>Paul > >> > >no I use QUOTA but put up with the performance hit, all my production > >servers are dual core now and new ones will be core 2 duos so I work > >round it with raw power, also using SATA hds instead of PATA. I dont > >use QUOTA on every single server tho, just the commercial web servers. > > > >Chris > > Thanks for your reply. > > That is odd as I used two Dual Core CPUS (4 in total) XEON 64 with 16 > GIG of RAM and SATA HD on the amd64 branch with a very fast Areca > controller and the hit was so bad that the system was not usable at > all. It was like going back to a 386 (or beyond that). > > Did you provide any mail services on that server? I have a hunch it > was the multiple instant access to many files that was causing the > slowdown (50-100 of small file reads and writes per second). If you > only provided FTP type access then that may not have been as big of > an issue as you may not have had as many concurrent file writes. > > Thanks > > Paul > > email/www/ftp all provided using a commercial control panel solution to manage it all, but the commercial servers are a mix of light to medium users and I dont think anyone is stressing the email systems a lot, they all do anti spam and anti virus as well on emails. I have one server that is very heavy usage but has QUOTA disabled as its dedicated to one domain only and not needed to use hd quotas. Chris