From owner-freebsd-stable Tue Nov 14 23:43:19 2000 Delivered-To: freebsd-stable@freebsd.org Received: from earth.backplane.com (placeholder-dcat-1076843399.broadbandoffice.net [64.47.83.135]) by hub.freebsd.org (Postfix) with ESMTP id 64ECC37B4CF for ; Tue, 14 Nov 2000 23:43:11 -0800 (PST) Received: (from dillon@localhost) by earth.backplane.com (8.11.1/8.9.3) id eAF7gXO53980; Tue, 14 Nov 2000 23:42:33 -0800 (PST) (envelope-from dillon) Date: Tue, 14 Nov 2000 23:42:33 -0800 (PST) From: Matt Dillon Message-Id: <200011150742.eAF7gXO53980@earth.backplane.com> To: Donn Miller Cc: Garrett Wollman , stable@FreeBSD.ORG, mckusick@mckusick.com Subject: Re: Any outstanding soft-updates or FFS bugs (matching this description)? References: <200011131924.OAA53671@khavrinen.lcs.mit.edu> <200011140302.eAE32sV43340@earth.backplane.com> <3A1212BC.F648B2E9@cvzoom.net> Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG : :Matt Dillon wrote: : :> I've had filesystems blow up under test when filling a parition :> with softupdates turned on. The most recent time was last :> weekend while I was testing low-memory stuff. : :It would be nice if the softupdates code (or maybe other kernel code :could do this) could detect when filesystem space is running low, and :kick into some "fail-safe" mode. For example, when the free space drops :below 15%, softupdates could automatically shut itself off, and then :turn itself back on when the free space rises above 20% (hysteresis). : :- Donn Mmm. It would be better to fix the bug, I think, then to try to program end-runs around it. I'm hoping that with all the reports coming in Kirk will be able to locate the problem (whether it be in softupdates or the buffer cache or whatever). A friend of mine just lost one of his filesystems to the same sort of corruption. -Matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message