From owner-freebsd-stable Thu Mar 6 21:56:32 2003 Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5AD3A37B405 for ; Thu, 6 Mar 2003 21:56:25 -0800 (PST) Received: from mobile.hub.org (u173n136.eastlink.ca [24.224.173.136]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3CE9643FD7 for ; Thu, 6 Mar 2003 21:56:23 -0800 (PST) (envelope-from scrappy@hub.org) Received: by mobile.hub.org (Postfix, from userid 1000) id C221D3F92; Fri, 7 Mar 2003 01:56:21 -0400 (AST) Received: from localhost (localhost [127.0.0.1]) by mobile.hub.org (Postfix) with ESMTP id B639C3F8B; Fri, 7 Mar 2003 01:56:21 -0400 (AST) Date: Fri, 7 Mar 2003 01:56:21 -0400 (AST) From: The Hermit Hacker X-X-Sender: scrappy@localhost To: Marc Spitzer Cc: freebsd-stable@FreeBSD.ORG Subject: Re: "leak" in softupdates? In-Reply-To: <20030306192939.3ec42f5e.mspitze1@optonline.net> Message-ID: <20030307014946.A40489@localhost> References: <20030305204526.T38115@hub.org> <200303060846.07649.darcy@wavefire.com> <20030306192939.3ec42f5e.mspitze1@optonline.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Thu, 6 Mar 2003, Marc Spitzer wrote: > On Thu, 06 Mar 2003 08:46:07 -0800 > Darcy Buskermolen wrote: > > > > > I have had simular experiances with this very phenominom under 3.X, > > although I have only ever seen it when you fill the partition ie 101% > > used and it only happens on partitions with softupdates. > > > > I do not mean to state the obvious, but have you tried to > look for open files on the partition, especially ones that have > been deleted but are still held open by the process? lsof will > give you a list. I am just mentioning it because it has got me > in the past. You could also try turning off daemon processes > one at a time and see if that helps. Tried that, using both lsof and fstat ... nothing is showing up ... in my case, I'm suspecting that the system is to busy for softupates to be able to sync to disk properly ... if I type 'reboot' to reboot the machine, it seems to take longer then normal to actually drop my connection, which I'm guessing is the time its taking to write everything to the drive ... I just disabled softupdates on that file system, so will see if that makes any difference ... And before you say anything about the 'too busy' comment, Tor Egge just spent the past week with me identifyng a "bug" in vnlru_proc where it couldn't reclaim vnodes fast enough, causing the server to hang ... he provided me with a patch to changes it reclaimation algorithm to speed things up to a reasonable level ... > > Good luck > > marc > > > > > On Wednesday 05 March 2003 16:53, Marc G. Fournier wrote: > > > I rebooted my server 20hrs ago, and had 20gig of free space ... now > > > I'm down to 4.6gig ... > > > > > > venus# df -t ufs > > > Filesystem 1K-blocks Used Avail Capacity Mounted on > > > /dev/amrd0s1a 103777974 90836532 4639206 95% / > > > > > > Before I rebooted it 20hrs ago, it had drop'd down <500Meg, and > > > after the reboot, pop'd back up to 20gig ... > > > > > > I can't seem to find where 16gig of disk space is being used though, > > > but, for instance, I had 12gig of files in /var/vmcore (from crash > > > dumps) that I removed, which should have increased me to 16gig free, > > > but I'm still at 4.6gig and drop'ng ... > > > > > > The system isn't particularly busy: > > > > > > venus# iostat 5 > > > tty amrd0 cpu > > > tin tout KB/t tps MB/s us ni sy in id > > > 1 116 0.00 0 0.00 32 0 28 1 40 > > > 1 452 4.29 343 1.43 39 0 52 0 9 > > > 0 225 5.21 258 1.31 32 0 47 1 21 > > > 2 35 6.21 270 1.64 13 0 51 0 36 > > > 1 38 4.46 268 1.17 19 0 49 1 31 > > > 1 32 4.81 261 1.23 16 0 51 1 32 > > > > > > And the drive is 7x18gig in a RAID5 configuration .. > > > > > > OS is 4.8-RC FreeBSD 4.8-RC #38: Tue Mar 4 22:43:18 CST 2003 with a > > > patch from Tor Egge to fix a problem with how vnodes are kept track > > > of and reclaimed/reused ... > > > > > > I've tried doing a 'sync', figuring that maybe it would force all > > > outstanding writes to the drive, but that seems to make no > > > difference either ... > > > > > > I'm doing a du right now to see if I can see anything large out of > > > the ordinary, but is there any way (maybe using lsof?) of finding > > > out what, if any, processes are holding open a large file? > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > > > with "unsubscribe freebsd-stable" in the body of the message > > > > -- > > Darcy Buskermolen > > Wavefire Technologies Corp. > > ph: 250.717.0200 > > fx: 250.763.1759 > > http://www.wavefire.com > > > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > > with "unsubscribe freebsd-stable" in the body of the message > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-stable" in the body of the message > Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy Systems Administrator @ hub.org primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message