From owner-cvs-all Mon Jan 29 1: 5:18 2001 Delivered-To: cvs-all@freebsd.org Received: from earth.backplane.com (earth-nat-cw.backplane.com [208.161.114.67]) by hub.freebsd.org (Postfix) with ESMTP id 8878D37B402; Mon, 29 Jan 2001 01:04:56 -0800 (PST) Received: (from dillon@localhost) by earth.backplane.com (8.11.1/8.9.3) id f0T93ms62313; Mon, 29 Jan 2001 01:03:48 -0800 (PST) (envelope-from dillon) Date: Mon, 29 Jan 2001 01:03:48 -0800 (PST) From: Matt Dillon Message-Id: <200101290903.f0T93ms62313@earth.backplane.com> To: Matthew Jacob Cc: Poul-Henning Kamp , John Baldwin , cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, Mikhail Teterin Subject: workaround for broken mfs using vn. (was Re: cvs commit: src/sys/i386/conf GENERIC ) References: Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG It's sad to see mfs not work, but I'm sure Poul will fixup whatever problems md has replacing it real quick. In the mean time, you can use 'vn' to workaround the MFS issue. It's a little messy, but it works: vnconfig -c -s labels,reserve -S 256m /dev/vn0 disklabel -r -w vn0c auto newfs /dev/vn0c tunefs -n enable /dev/vn0c mount /dev/vn0 /usr/src/sys/compile (or whatever) Works for me w/ the latest -current source. Be careful not eat up all your swap, though. I recommend always using the 'reserve' keyword even though it eats more swap. As a bonus, if you want to try to recover the VN partition after a reboot (but not after a crashdump), you can try fsck'ing it rather then newfs'ing. Try doing that with the old mfs! vnconfig -c -s labels,reserve -S 256m /dev/vn0 disklabel -r -w vn0c auto fsck /dev/vn0c if [ $? != 0 ]; then newfs /dev/vn0c tunefs -n enable /dev/vn0c fi mount /dev/vn0 /usr/src/sys/compile (or whatever) -Matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message