From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <3E95CCDB.4080305@cox.net> From: "Kevin P. Fleming" MIME-Version: 1.0 Subject: Re: [linux-lvm] lvm/devfs bug? References: <1049829705.2312.16.camel@rich> <20030408204334.GD22951@percy.comedia.it> <1049836943.7411.19.camel@rich> <20030409072130.GD1366@tykepenguin.com> <1049932652.7946.4.camel@rich-laptop> <1049998095.1031.5.camel@rich-laptop> In-Reply-To: <1049998095.1031.5.camel@rich-laptop> Content-Transfer-Encoding: 7bit Sender: linux-lvm-admin@sistina.com Errors-To: linux-lvm-admin@sistina.com Reply-To: linux-lvm@sistina.com List-Help: List-Post: List-Subscribe: , List-Unsubscribe: , List-Archive: Date: Thu Apr 10 14:58:01 2003 List-Id: Content-Type: text/plain; charset="us-ascii"; format="flowed" To: linux-lvm@sistina.com rich turner wrote: > does anyone recognize this as a bug within lvm? > > there clearly is not a partition on sda (according to /proc/partitions) > but pvcreate thinks there is one. if pvcreate examines only > /proc/partitions for disk partitions and there is not a partition on a > disk then why does pvcreate fail? i may not understand all of the > underlying behavior of pvcreate but it appears to be failing when it > shouldnt. is this a bug? > If I remember your original message properly, the error you got from pvcreate was that the disk "contained a partition table". This is _not_ the same thing as the disk containing one or more partitions; there can very well be an empty partition table on the disk. This may not be the problem, but if it is, using dd if=/dev/zero of=/dev/sda count=20 will wipe out any existing DOS-type partition table (or any other type of table that resides at the beginning of the disk). If you've already tried this, then I'd say there is definitely a bug happening.