On Thu, Oct 23, 2014 at 02:41:47AM +0000, Duncan wrote: > Dave posted on Wed, 22 Oct 2014 08:49:46 -0400 as excerpted: > > > On Tue, Oct 21, 2014 at 10:08 PM, Duncan <1i5t5.dun...@cox.net> wrote: > >> As for the mounted filesystem question, since all it does is flip a > >> switch so that new metadata writes use the skinny-metadata code path, > >> it shouldn't be a problem. > > > > Nope. Just tried it here: > > > > # btrfs --version Btrfs v3.16.1-42-g140eccb > > > > # btrfstune -x /dev/dm-0 /dev/dm-0 is mounted > > Thanks. > > So btrfstune refuses to set the skinny-metadata flag at all on mounted > devices. Nicely reduces risk, /and/ answers the question. =:^)
btrfstune requires an unmounted device. The on-line change to features is done via the sysfs interface, eg /sys/fs/btrfs/<UUID>/features, then echo 1 > featurename. Right now only the extended refs (aka hardlink limit) can be turned on. -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
