[systemd-devel] fstab, rootfs on btrfs

[Chris Murphy]

In Fedora 20, by default anaconda sets fs_passno in fstab to 1 for / on btrfs. During offline updates, this is causing systemd-fstab-generator to freak out not finding fsck.btrfs.

https://bugzilla.redhat.com/show_bug.cgi?id=1034563

For some time I've been suggesting that fstab should use fs_passno 0 for btrfs file systems:
https://bugzilla.redhat.com/show_bug.cgi?id=862871

But because of this suggestion by an XFS dev, I'm wondering if that's not a good idea. Or if we should expect some smarter behavior on the part of systemd (now or in the future) when it comes to devices that take a long time to appear?
http://www.mail-archive.com/linux-btrfs@vger.kernel.org/msg29231.html

It doesn't seem to me that for file systems that don't require an fs check, that fstab should indicate it does require an fs check, just to inhibit hissy fits by other processes not liking that the device is missing. But maybe I'm missing something.


Chris Murphy