From owner-freebsd-current Mon Feb 10 3:20: 9 2003 Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9942037B40A for ; Mon, 10 Feb 2003 03:20:07 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3729143FA3 for ; Mon, 10 Feb 2003 03:20:06 -0800 (PST) (envelope-from j@uriah.heep.sax.de) Received: (from uucp@localhost) by sax.sax.de (8.9.3/8.9.3) with UUCP id MAA25087 for current@freebsd.org; Mon, 10 Feb 2003 12:20:05 +0100 (CET) Received: from uriah.heep.sax.de (localhost.heep.sax.de [127.0.0.1]) by uriah.heep.sax.de (8.12.6/8.12.6) with ESMTP id h1ABDl2X010572 for ; Mon, 10 Feb 2003 12:13:47 +0100 (MET) (envelope-from j@uriah.heep.sax.de) Received: (from j@localhost) by uriah.heep.sax.de (8.12.6/8.12.6/Submit) id h1ABDlDw010571 for current@freebsd.org; Mon, 10 Feb 2003 12:13:47 +0100 (MET) Date: Mon, 10 Feb 2003 12:13:47 +0100 From: Joerg Wunsch To: current@freebsd.org Subject: Abysmal performance for restore -r Message-ID: <20030210121347.A8577@uriah.heep.sax.de> Reply-To: Joerg Wunsch Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Since growfs still doesn't work, i needed to newfs & restore the following filesystem after expanding the volume: uriah # df -k -i /dev/vinum/home_cvs Filesystem 1K-blocks Used Avail Capacity iused ifree %iused Mounted on /dev/vinum/home_cvs 4125838 2911856 883916 77% 235970 305724 44% /home/cvs restore(8)ing it from tape took about 7.5 hours(!) in order to create the 34332 directory inodes on it; that's about 1.2 directory creations per second. The machine was basically grinding to a halt while this happened, iostat (when it was possible to report something at all) reported > 350 transactions per second for both participating disks (which doesn't sound all that bad to me, the disks are 10000 rpm not too old drives). In contrast, restoring the ~ 3 GB of contents (still > 250000 files) didn't even take an hour to complete. The filesystem has been newfs'ed with defaults + softupdates: uriah # newfs -N -U /dev/vinum/home_cvs /dev/vinum/home_cvs: 4096.0MB (8388608 sectors) block size 16384, fragment size 2048 using 23 cylinder groups of 183.69MB, 11756 blks, 23552 inodes. with soft updates Does anybody have an idea why this was so terrible slow? (The kernel is at 5.0-RCsomething, in case that matters.) -- cheers, J"org .-.-. --... ...-- -.. . DL8DTL http://www.sax.de/~joerg/ NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message