On Fri, Jul 10, 2020 at 12:16 PM Lennart Poettering <mzerqung@xxxxxxxxxxx> wrote: > > On Fr, 10.07.20 09:34, Chris Murphy (lists@xxxxxxxxxxxxxxxxx) wrote: > > > > That makes things a lot more robust, as btrfs will then just work like > > > any other fs even if you insert the root subvol in between like > > > anaconda apparently does. > > > > > > I think there's big value in allowing short kernel cmdlines that are > > > as similar as possible everywhere, instead of blowing it up with > > > different switches for every single case. > > > > I agree with all of the above, but there is a contra argument. There is > > something to be said about having an understandable system, one that self > > describes how it's assembled, and boots. Changing the default subvolume > > obscures this, and now one of the "connect the dots" steps of boot becomes > > a dot on a completely different page in another book. > > I am pretty sure a self-describing system is the best system. And if > one subvolume is marked "as default" and then booted into "as default" > then that's fantastically self-descriptive. > > The concept of default subvolumes exists, for cases like this, and it > simplifies things and makes them more robust. > > There's really no need to complicate things by pushing btrfsisms into > user-visible concepts needlessly. It's been this way in Fedora for a decade. The behavior is baked into the installer, and bootloader scripts. The 'rootflags' boot parameter is added by the grub-mkconfig scripts. It's upstreamed. When I say understandable, I mean a user can follow a trail of breadcrumbs for boot, startup, and assembly: not just rootflags, but also that systemd at sysroot mount time, also uses the 'subvol=' mount option. Whereas if this is a leap, they can't follow it. If the user mounts this file system elsewhere, including from a livecd, they actually won't see their home, it's hidden. How are they supposed to know where it is or how to find it, without learning btrfsisms? I've seen a lot more confusion from the (open)SUSE use of set-default than on Fedora where it isn't used. *shrug* I agree Fedora way is more verbose. But that also makes it easier to understand and troubleshoot and *not* have to know about decoder rings like "btrfs subvolume get-default" in order to find out why some subvolume is always mounted, the others are all invisible. This all relates to a possible snapshotting and rollback feature too. And that's going to take design and planning, including Anaconda, bootloader, and desktop teams being onboard. I'm disinclined to add churn almost everywhere else, just to eliminate one boot parameter. -- Chris Murphy _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
