On Fri, Sep 9, 2016 at 12:58 PM, Chris Murphy <li...@colorremedies.com> wrote:
> > It should work better than it does because it works well for LVM and > mdadm arrays. > > I think what's going on is the DE's mounter (udisksd) tries to mount > each Btrfs device node, even though those nodes make up a single fs > volume. I updated this bug but it looks like it's going to a maintainer whose not reading these mails. https://bugs.freedesktop.org/show_bug.cgi?id=87277#c3 Anyway the problem is pretty bad as I describe in that bug. It pretty much will always cause some kind of Btrfs corruption, which does get fixed if everything is raid1. The main flaw is that it uses sysfs to delete the device node before umounting the file system, so it causes any multiple device Btrfs array that's automounted to become degraded. There's something of a silver lining so long as the metadata at least is raid1, which it probably ought to be, rather than single, dup, or raid0. In any case, it's far worse and more dangerous than LVM or mdadm raid which don't exhibit this behavior. So of the available short term options I see, udisks2 (and now storaged) needs to black list Btrfs from automounting. It's just not smart enough to make sure it's not doing really wrong things. -- Chris Murphy -- 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
