From owner-freebsd-scsi Sun Apr 9 3:54:46 2000 Delivered-To: freebsd-scsi@freebsd.org Received: from news.lucky.net (news.lucky.net [193.193.193.102]) by hub.freebsd.org (Postfix) with ESMTP id 403DF37B6F3 for ; Sun, 9 Apr 2000 03:51:36 -0700 (PDT) (envelope-from vovik@burka.carrier.kiev.ua) Received: (from mail@localhost) by news.lucky.net (8.Who.Cares/8.Who.Cares) id NTI19077 for freebsd-scsi@freebsd.org; Sun, 9 Apr 2000 13:50:12 +0300 (envelope-from vovik@burka.carrier.kiev.ua) From: "Vladimir A. Jakovenko" To: freebsd-scsi@freebsd.org Subject: CAM access from userland Date: 9 Apr 2000 10:48:31 GMT Organization: Lucky Net, Kiev Message-ID: <8cpn5v$ija$1@news.lucky.net> X-Trace: news.lucky.net 955277311 19050 193.193.193.107 (9 Apr 2000 10:48:31 GMT) X-Complaints-To: usenet@news.lucky.net User-Agent: tin/1.4.1-19991201 ("Polish") (UNIX) (FreeBSD/3.4-STABLE (i386)) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 13 Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi! I need to write userland programm which talks to external SCSI device (sattelite DVB modem) using nonstandart SCSI commands. Should I use cam(3)? Or should I use some sort of pass(4) or xpt(4) IOCTLs? Where I can find more documentation of filling CAM CCBes? -- Regards, Vladimir. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message From owner-freebsd-scsi Sun Apr 9 4:17:16 2000 Delivered-To: freebsd-scsi@freebsd.org Received: from relay02.chello.nl (relay02.chello.nl [212.83.68.146]) by hub.freebsd.org (Postfix) with ESMTP id 38C5437B55E for ; Sun, 9 Apr 2000 04:16:53 -0700 (PDT) (envelope-from wkb@chello.nl) Received: from chello.nl ([213.46.78.184]) by relay02.chello.nl (InterMail vK.4.02.00.00 201-232-116 license 99c8f334c649856e3f2cdadc4054e412) with ESMTP id <20000409111434.MGPB23338.relay02@chello.nl>; Sun, 9 Apr 2000 13:14:34 +0200 Received: (from wkb@localhost) by chello.nl (8.9.3/8.9.3) id NAA47641; Sun, 9 Apr 2000 13:14:53 +0200 (CEST) (envelope-from wkb) Date: Sun, 9 Apr 2000 13:14:53 +0200 From: Wilko Bulte To: "Vladimir A. Jakovenko" Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: CAM access from userland Message-ID: <20000409131453.A47627@yedi.wbnet> Reply-To: wc.bulte@chello.nl References: <8cpn5v$ija$1@news.lucky.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <8cpn5v$ija$1@news.lucky.net>; from vovik@burka.carrier.kiev.ua on Sun, Apr 09, 2000 at 10:48:31AM +0000 X-OS: FreeBSD 3.4-STABLE X-PGP: finger wilko@freebsd.org Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Apr 09, 2000 at 10:48:31AM +0000, Vladimir A. Jakovenko wrote: > Hi! > > I need to write userland programm which talks to external SCSI device > (sattelite DVB modem) using nonstandart SCSI commands. > > Should I use cam(3)? Or should I use some sort of pass(4) or xpt(4) IOCTLs? > > Where I can find more documentation of filling CAM CCBes? man camcontrol -- Wilko Bulte Powered by FreeBSD http://www.freebsd.org http://www.tcja.nl To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message From owner-freebsd-scsi Wed Apr 12 12: 5:19 2000 Delivered-To: freebsd-scsi@freebsd.org Received: from darcy.gwis.com (darcy.gwis.com [209.57.72.3]) by hub.freebsd.org (Postfix) with ESMTP id 6BEBA37C157 for ; Wed, 12 Apr 2000 12:00:57 -0700 (PDT) (envelope-from ddr@gwis.com) Received: from localhost (ddr@localhost) by darcy.gwis.com (8.9.3/8.9.3) with ESMTP id OAA85758 for ; Wed, 12 Apr 2000 14:57:44 -0400 (EDT) Date: Wed, 12 Apr 2000 14:57:44 -0400 (EDT) From: Dan Roberts - GWIS To: freebsd-scsi@freebsd.org Subject: Possible CAM scheduling problem with DPT SrIV controller? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I've got a SmartRAID IV controller in my FreeBSD 3.4 machine.. when disk activity becomes high, the system is unresponsive. I've spoken with the driver's author and it's his opinion that the problem lies withing the SCSI CAM subsystem. I can demonstrate this problem easily by making the system work on a large file, as demonstrated below. $ dd if=/dev/zero of=/home/ddr/bigfile bs=1k count=200000 This is a snapshot of the system under a lighter than average load, before issuing the dd command (increments of 1 second): tty da0 cpu tin tout KB/t tps MB/s us ni sy in id 12 240 6.83 30 0.20 12 0 8 1 79 15 2039 7.36 11 0.08 9 0 12 1 78 18 325 7.29 57 0.41 13 0 12 3 72 18 2074 6.76 85 0.56 19 0 42 2 37 12 412 7.32 19 0.13 3 0 10 1 86 89 2234 6.75 16 0.10 8 0 13 2 78 This is the same system a few minutes later while dd is running: tty da0 cpu tin tout KB/t tps MB/s us ni sy in id 0 455 16.73 184 3.01 2 0 1 2 95 0 688 8.83 325 2.80 2 0 2 2 95 0 229 21.73 65 1.39 8 0 53 2 38 0 2080 64.00 30 1.86 1 0 2 2 95 0 229 14.28 212 2.95 2 0 0 3 95 0 1973 13.07 203 2.59 9 0 43 2 47 0 229 64.00 49 3.03 6 0 30 0 64 0 2029 13.55 154 2.04 2 0 2 2 94 If you're using the system while the dd process is running, you'll find that anything you have open at the moment that doesn't require disk access continues to run fine. However, as soon as vi needs to access the file you're working on, or you call a new program from the shell prompt, the session freezes until dd is complete. Is there anyone familiar enough with this kind of problem that can offer me any advise? I'm afraid this is will over my head. -- Dan Roberts, Systems Engineer Voice 800.656.GWIS GWIS Internet Solutions Fax 330.656.5440 Please contact support@gwis.com for all technical support issues To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message From owner-freebsd-scsi Wed Apr 12 13:36:51 2000 Delivered-To: freebsd-scsi@freebsd.org Received: from aurora.sol.net (aurora.sol.net [206.55.65.76]) by hub.freebsd.org (Postfix) with ESMTP id 3613037B5A8 for ; Wed, 12 Apr 2000 13:36:48 -0700 (PDT) (envelope-from jgreco@aurora.sol.net) Received: (from jgreco@localhost) by aurora.sol.net (8.9.2/8.9.2/SNNS-1.02) id LAA03450; Sat, 15 Apr 2000 11:40:25 -0500 (CDT) From: Joe Greco Message-Id: <200004151640.LAA03450@aurora.sol.net> Subject: Re: hardware vs software stripping To: grog@lemis.com, scsi@FreeBSD.ORG Date: Sat, 15 Apr 2000 11:40:25 -0500 (CDT) X-Mailer: ELM [version 2.4ME+ PL43 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I think one of the problems is that I can't find an authoritative > definition of the levels. I was going to buy one of those > super-expensive books that you probably have, but in the meantime I've > been limited to various web pages. I'm reasonably certain that one could trust the original Berkeley paper on RAID. http://sunsite.berkeley.edu:80/Dienst/UI/2.0/ShowPage/ncstrl.ucb/CSD-87-391?npages=26&format=inline&page=1 If that isn't an authoritative, non-marketeered definition of the levels, we might as well just all go home and define what the levels mean ourselves. -- ... Joe ------------------------------------------------------------------------------- Joe Greco - Systems Administrator jgreco@ns.sol.net Solaria Public Access UNIX - Milwaukee, WI 414/342-4847 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message From owner-freebsd-scsi Thu Apr 13 8:43:38 2000 Delivered-To: freebsd-scsi@freebsd.org Received: from privatecube.privatelabs.com (silvercube.silverpix.com [198.143.31.30]) by hub.freebsd.org (Postfix) with ESMTP id 2282C37B9F6; Thu, 13 Apr 2000 08:43:35 -0700 (PDT) (envelope-from mi@privatelabs.com) Received: from misha.privatelabs.com (root@misha.privatelabs.com [198.143.31.6]) by privatecube.privatelabs.com (8.9.2/8.9.2) with ESMTP id LAA22853; Thu, 13 Apr 2000 11:42:25 -0400 (EDT) Received: from privatelabs.com (mi@localhost [127.0.0.1]) by misha.privatelabs.com (8.9.3/8.9.3) with ESMTP id LAA14311; Thu, 13 Apr 2000 11:43:17 -0400 (EDT) (envelope-from mi@privatelabs.com) From: mi@privatelabs.com Message-Id: <200004131543.LAA14311@misha.privatelabs.com> Date: Thu, 13 Apr 2000 11:43:15 -0400 (EDT) Subject: configuring squid To: questions@freebsd.org, scsi@freebsd.org MIME-Version: 1.0 Content-Type: TEXT/plain; CHARSET=US-ASCII Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello! I'm setting up a fairly big squid server with two 45Gb (but slow) SCSI SEAGATE ST446452W (external). I wonder if I should use ccd to make one 90Gb interleaved array of them or use them separately and tell Squid about the two independent partitions... Speed is the only factor -- I understand, that separately they'd be easier to manage... This is FreeBSD-4-STABLE... Any suggestions? TIA! -mi To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message From owner-freebsd-scsi Thu Apr 13 11:37:13 2000 Delivered-To: freebsd-scsi@freebsd.org Received: from mail.wolves.k12.mo.us (mail.wolves.k12.mo.us [207.160.214.1]) by hub.freebsd.org (Postfix) with ESMTP id BE5EA37B689; Thu, 13 Apr 2000 11:37:08 -0700 (PDT) (envelope-from cdillon@wolves.k12.mo.us) Received: from mail.wolves.k12.mo.us (cdillon@mail.wolves.k12.mo.us [207.160.214.1]) by mail.wolves.k12.mo.us (8.9.3/8.9.3) with ESMTP id NAA76572; Thu, 13 Apr 2000 13:36:56 -0500 (CDT) (envelope-from cdillon@wolves.k12.mo.us) Date: Thu, 13 Apr 2000 13:36:56 -0500 (CDT) From: Chris Dillon To: mi@privatelabs.com Cc: questions@FreeBSD.ORG, scsi@FreeBSD.ORG Subject: Re: configuring squid In-Reply-To: <200004131543.LAA14311@misha.privatelabs.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Snipped from freebsd-scsi, this is only appropriate for freebsd-questions. On Thu, 13 Apr 2000 mi@privatelabs.com wrote: > Hello! I'm setting up a fairly big squid server with two 45Gb (but slow) > SCSI SEAGATE ST446452W (external). > > I wonder if I should use ccd to make one 90Gb interleaved array of them > or use them separately and tell Squid about the two independent > partitions... Speed is the only factor -- I understand, that separately > they'd be easier to manage... Keep them separate. Squid load-balances among multiple cache_dirs. If speed is the biggest factor, you really should be using many smaller drives with a single cache_dir on each one, instead of two large drives. Keep in mind you're also going to need a lot of memory for a full 90GB cache. You need at least 10MB RAM per 1GB of cache (this is from my personal experience with Squid, and does not include OS overhead, filesystem cache, or anything else), so you'll need at least 1GB in there. Since you're also going to be using two large disks instead of many smaller ones, you'll want plenty of RAM available for the filesystem cache and to increase Squid's cache_mem significantly above the default of 8MB to hold the most popular objects without having to fetch them from disk often. How many requests per second are you expecting during peak times, anyway? What you consider "fairly big" could in fact be humungous, or it could be just a drop in the bucket. Knowing this would help determine wether what you have will be enough, or complete overkill. -- Chris Dillon - cdillon@wolves.k12.mo.us - cdillon@inter-linc.net FreeBSD: The fastest and most stable server OS on the planet. For Intel x86 and Alpha architectures. ( http://www.freebsd.org ) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message