On systems with SELinux enabled, we have to create the directories on
top of which we mount another partition or virtual file system (e.g.,
/dev) with the correct SELinux context, BEFORE we mount the other
partition. Otherwise, SELinux will get really confused when systemd
tries to recreate the mount tree for a private file system namespace for
a service. And unfortunately, even an autorelabel does not fix it
because it runs when /dev etc. are already mounted.
Without this fix, on Fedora >= 30, the system installed with Calamares
would fail to start the dbus-broker system bus, leading to several
important pieces of functionality not working (e.g., shutdown as
non-root).
On systems without SELinux enabled, chcon (which is part of coreutils)
will just print a warning and do nothing, so this should always be safe.
- One might argue whether an empty list of partitions to mount is
a bad thing. It suggests that the partition module wasn't used,
and so we're in an OEM situation -- but then everything should
already be mounted anyway. That's why I choose empty -> bail.
- If a key is missing from mount.conf, don't raise KeyError
- If both keys are missing, suggest that mount.conf might
be missing instead (a consequence of INSTALL_CONFIG=OFF, for
instance).
- Simplify code a bit.
- Don't bother returning None explicitly.
Only the string "fat32" and "fat16" will be replaced with vfat. If an
case sensitive "Fat32" some problems occure:
- mount: partition cannot be mounted (e.g. a fat32 efi partition)
- fstab: system won't even boot because fstab does not know the type "Fat32"
mount: Remember the extraMounts (and extraMountsEfi, if applicable) in
the global storage (as "extraMounts").
unpackfs: Read the extra mounts from the global storage to generate the
exclude list instead of trying to detect it from the "mount" command's
output, because the latter also includes normally-mounted destination
partitions.
This makes having separate partitions for non-/ mountpoints work again.
It turns out that I had messed this up. Issue #173 was addressed anyway
by the followup fix to unpackfs to only exclude the directories that are
actually mounted (which does work), but with this fix, /run is now
handled as intended.
PS: Should the mount job not check the return value of
libcalamares.utils.mount instead of failing silently?
Module descriptors are now module.desc, no configuration allowed inside.
Module config files are <modulename>.conf, installed in
share/calamares/modules.
settings.conf is read from /etc/calamares, then share/calamares, and if
running with --debug also in the current dir.
Module config files are read from /etc/calamares/modules, then
share/calamares/modules, and if running with --debug also in
src/modules/<modulename> relative to the current dir.
