From owner-freebsd-hackers Tue Sep 21 5:12: 3 1999 Delivered-To: freebsd-hackers@freebsd.org Received: from caxton.correionet.com.br (caxton.correionet.com.br [200.246.35.7]) by hub.freebsd.org (Postfix) with ESMTP id 10E9B151F4 for ; Tue, 21 Sep 1999 05:11:39 -0700 (PDT) (envelope-from morte@correionet.com.br) Received: from caxton.correionet.com.br (caxton.correionet.com.br [200.246.35.7]) by caxton.correionet.com.br (8.9.3/8.9.3) with ESMTP id JAA03942 for ; Tue, 21 Sep 1999 09:10:45 -0300 (EST) Date: Tue, 21 Sep 1999 09:10:44 -0300 (EST) From: Luiz Morte da Costa Junior To: freebsd-hackers@freebsd.org Subject: Re: L440GX+ Server Board In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hi list, My problem is solved. Since I changed my SCSI to a Quantum, I haven't have more problem with performance. I recompiled the pwd_mkdb.c program using more RAM memory and the performance became better. Before compilation the "delete user" operation takes 3 minutes, and now the same operation takes 11 seconds (whe have about 12000 usernames - thanks Pedro Vazquez ;). []s, Luiz Morte da Costa Junior Analista de Redes E-mail: morte@correionet.com.br Telefone: +55 19 754-2532 Fax: +55 19 255-7576 CorreioNet - Correio Popular Campinas - SP - Brazil On Mon, 13 Sep 1999, Luiz Morte da Costa Junior wrote: > > Hi list, > > I can't solve the problem yet. > > When I runnig the dmesg command, I have: > > --- > da0 at ahc0 bus 0 target 0 lun 0 > da0: Fixed Direct Access SCSI-2 device > da0: 40.000MB/s transfers (20.000MHz, offset 15, 16bit), Tagged Queueing > Enabled > da0: 8683MB (17783240 512 byte sectors: 255H 63S/T 1106C) > da1 at ahc0 bus 0 target 1 lun 0 > da1: Fixed Direct Access SCSI-2 device > da1: 40.000MB/s transfers (20.000MHz, offset 15, 16bit), Tagged Queueing > Enabled > da1: 8683MB (17783240 512 byte sectors: 255H 63S/T 1106C) > --- > > I did a test, changing only the SCSI disk, using the same motherboard > (L440GX+). I ran the dmesg command again and I received: > > -- > da0 at ahc0 bus 0 target 0 lun 0 > da0: Fixed Direct Access SCSI-2 device > da0: 80.000MB/s transfers (40.000MHz, offset 31, 16bit), Tagged Queueing > Enabled > da0: 8683MB (17783250 512 byte sectors: 255H 63S/T 1106C) > -- > > > I realized that the quantum disk have a 80.000MB/s transfers. > Does someone use a SEAGATE ST39140LW disk and have I/O problems? > > Regards, > Luiz Morte. > > PS: > Someone told me that vipw doens't have much I/O disk, but when I run it > with 12000 accounts, my disk have a lot off access. > > On Sun, 22 Aug 1999, Luiz Morte da Costa Junior wrote: > > > > Hi all, > > > > I have a problem with a server running a FreeBSD 3.2. > > > > My Server has a L440GX+ Serber Board (intel), with network card 10/100, > > SCSI AIC 7896 (80MB/s), 2 SCSI disk with 9GB (80MB/s), 2 pentium III > > 450Mhz (not overclocked). The NIC and SCSI are onboard. > > > > I recompiled a kernel to SMP, and it worked. The server is ok, but when > I > > run a comand with disk access (whith vipw or mysql), the performance of > > server goes down. My server stays very very very slow. If I use pine to > > read my messages, it doesn't work. When the comand finishes, the server > > stays "ok" again. > > > > I recompiled the kernel with "maxusers 128", but it doesn't work. > > > > My SCSI cable has a terminator and the scsi setup is ok (I think :) ). > > > > The dmesg command output is in attchmnt. > > > > I appreciate any help. > > []s, > > Luiz Morte da Costa Junior > Analista de Redes E-mail: morte@correionet.com.br > Telefone: +55 19 754-2532 Fax: +55 19 255-7576 > CorreioNet - Correio Popular Campinas - SP - Brazil > []s, Luiz Morte da Costa Junior Analista de Redes E-mail: morte@correionet.com.br Telefone: +55 19 754-2532 Fax: +55 19 255-7576 CorreioNet - Correio Popular Campinas - SP - Brazil To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message