On Mon, Nov 30, 2009 at 12:41:29PM +0100, Leszek Ciesielski wrote: > Hi, > > after adding a second device to a btrfs filesystem (kernel 2.6.31) > used for root fs, I am getting a "No filesystem could mount root" > message upon reboot. Is the caveat > > "btrfsctl -a is used to scan all of the block devices under /dev and > probe for Btrfs volumes. This is required after loading the btrfs > module if you're running with more than one device in a filesystem. " > > still true? If yes, is there a technical reason for it - am I wrong to > assume that btrfs could execute the scan upon mount request when the > filesystem requires it for assembling? > The btrfs module needs to have an idea of what devices are available for it to use. The btrfsctl -a needs to be done in userspace since it scans /dev for block devices to probe to see if they are btrfs devices. To do it in the kernel would require calling out to userspace to run btrfsctl -a for the uncommon case that you are booting from a multi-disk setup. Hopefully as btrfs becomes more and more common distributions will start putting btrfsctl -a in their initrd's if they detect you have a mutli-disk root. Thanks, Josef -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
