From owner-freebsd-stable@FreeBSD.ORG Tue Feb 6 19:44:05 2007 Return-Path: X-Original-To: freebsd-stable@freebsd.org Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id CE1F916A402 for ; Tue, 6 Feb 2007 19:44:05 +0000 (UTC) (envelope-from matrix@itlegion.ru) Received: from corpmail.itlegion.ru (corpmail.itlegion.ru [84.21.226.211]) by mx1.freebsd.org (Postfix) with SMTP id F156713C461 for ; Tue, 6 Feb 2007 19:44:04 +0000 (UTC) (envelope-from matrix@itlegion.ru) Received: (qmail 82903 invoked from network); 6 Feb 2007 22:44:00 +0300 Received: from unknown (HELO Artem) (192.168.0.12) by 84.21.226.211 with SMTP; 6 Feb 2007 22:44:00 +0300 X-AntiVirus: Checked by Dr.Web [version: 4.33, engine: 4.33.5.10110, virus records: 173786, updated: 6.02.2007] Message-ID: <001d01c74a27$25afd9d0$0c00a8c0@Artem> From: "Artem Kuchin" To: Date: Tue, 6 Feb 2007 22:43:51 +0300 Organization: IT Legion MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset="koi8-r"; reply-type=original Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.3028 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3028 Subject: Filesystem corruption and bad perfomance with SRCS16 and PAE (raid 5 2TB) X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 06 Feb 2007 19:44:05 -0000 Hello! I've got an Intel 2400A server to install FBSD. It has 4GB RAM, SRCS16 raid controller, configured 2TB (actually a little less, about 1.9TB) RAID-5 array, Xeon 3.2Ghz Recently i have installed 6.2-R, cvsupped it to 6.2-STABLE (today, thanks to included csup utility, very convinient) and decided to turn on PAE so i can utilize all 4GB of RAM (w/o PAE i am loosing 512MB, too bad). I did that and then tried to run perl. Got "segmantaion fault". Ignored that and tried to copy port tree and then i found out that some files are damanged and connot be read. Also, directoried could not be deleted because it is not empty, but when i do ls in that dir it IS empty. weird. Booted single user, did fsck - found a buch of errors, many softupdate inconsistencies. Fixed them. Rebooted (still in PAE). Run buildworld and copy ports tree at the same time. AFter 10 minutes again the same errors. Built a kernel without PAE, installed, rebooted in single user, fixed all filesystems, booted multiuser, run buildworld and copy of ports tree. Everything is okay. Did it many times. Still everything is ok. Is it something bad about amr driver working under PAE? Also, with and after running in PAE (i mean, run PAE kernel, then install normal kernel and run it) the kernel compains like this _vfs_done(): amrd0 [WRITE(offset=8192, length=1536)] error = 5 (no other message before or aftrer this, disks are fully okay becuase dd works like a charm) Also, i did a simple thing dd if=/dev/amrd0 of=/dev/null bs=1M (tried 2m, 5m, 512k) The throughoutput is about 29MB/sec w/o PAE and 27MB/sec with PAE Did the same on a very old DUal Pentium III with 3Ware 7xxxx with RAID-5 array and older harddrives. Got over 50Mb/sec. unixbench show terrible number for file perfomance. Why the heck amr is so slow? -- Regards, Artem Kuchin