It appears LOOP_CONFIGURE in 5.8 is even more broken than initially
thought: it doesn't properly propgate lo_sizelimit to the block device
layer. :-(
Let's hence check the block device size immediately after issuing
LOOP_CONFIGURE, and if it doesn't match what we just set let's fallback
to the old ioctls.
This means LOOP_CONFIGURE currently works correctly only for the most
simply case: no partition table logic and no size limit. Sad!
(Kernel people should really be told about the concepts of tests and
even CI, one day!)
This matches how we handle things everywhere else, i.e. in .mount units,
and similar: when a mount point dir is missing, we create it, let's do
so too when dealing with disk images.
This makes things a lot simpler, more robust, and systematic.
fat is a bit more limited in volume name length and UUID support. Let's
add some special support for it.
This is particularly useful to generate EFI system partitions.
Kernel 5.8 gained a hidepid= implementation that is truly per procfs,
which allows us to mount a distinct once into every unit, with
individual hidepid= settings. Let's expose this via two new settings:
ProtectProc= (wrapping hidpid=) and ProcSubset= (wrapping subset=).
Replaces: #11670
We return BUS_ERROR_NO_SUCH_UNIT a.k.a. org.freedesktop.systemd1.NoSuchUnit
in various places. In #16813:
Aug 22 06:14:48 core sudo[2769199]: pam_systemd_home(sudo:account): Failed to query user record: Unit dbus-org.freedesktop.home1.service not found.
Aug 22 06:14:48 core dbus-daemon[5311]: [system] Activation via systemd failed for unit 'dbus-org.freedesktop.home1.service': Unit dbus-org.freedesktop.home1.service not found.
Aug 22 06:14:48 core dbus-daemon[5311]: [system] Activating via systemd: service name='org.freedesktop.home1' unit='dbus-org.freedesktop.home1.service' requested by ':1.6564' (uid=0 pid=2769199 comm="sudo su ")
This particular error comes from bus_unit_validate_load_state() in pid1:
case UNIT_NOT_FOUND:
return sd_bus_error_setf(error, BUS_ERROR_NO_SUCH_UNIT, "Unit %s not found.", u->id);
It seems possible that we should return a different error, but it doesn't really
matter: if we change pid1 to return a different error, we still need to handle
BUS_ERROR_NO_SUCH_UNIT as in this patch to handle pid1 with current code.
This patch adds seccomp support to the riscv64 architecture. seccomp
support is available in the riscv64 kernel since version 5.5, and it
has just been added to the libseccomp library.
riscv64 uses generic syscalls like aarch64, so I used that architecture
as a reference to find which code has to be modified.
With this patch, the testsuite passes successfully, including the
test-seccomp test. The system boots and works fine with kernel 5.4 (i.e.
without seccomp support) and kernel 5.5 (i.e. with seccomp support). I
have also verified that the "SystemCallFilter=~socket" option prevents a
service to use the ping utility when running on kernel 5.5.
On new kernels (>= 5.8) unprivileged users may create the 0:0 character
device node. Which is great, as we can use that as inaccessible device
nodes if we run unprivileged. Hence, change how we find the right
inaccessible device inodes: when the user asks for a block device node,
but we have none, try the char device node first. If that doesn't exist,
fall back to the socket node as before.
This means that:
1. in the best case we'll return a node if the right device node type
2. otherwise we hopefully at least can return a device node if one asked
for even if the type doesn't match (i.e. we return char instead of
the requested block device node)
3. in the worst case (old kernels…) we'll return a socket node
Similarly to "setup" vs. "set up", "fallback" is a noun, and "fall back"
is the verb. (This is pretty clear when we construct a sentence in the
present continous: "we are falling back" not "we are fallbacking").
Follow the same model established for RootImage and RootImageOptions,
and allow to either append a single list of options or tuples of
partition_number:options.
The concept is flawed, and mostly useless. Let's finally remove it.
It has been deprecated since 90a2ec10f2 (6
years ago) and we started to warn since
55dadc5c57 (1.5 years ago).
Let's get rid of it altogether.
Let's make /run/host the sole place we pass stuff from host to container
in and place the "inaccessible" nodes in /run/host too.
In contrast to the previous two commits this is a minor compat break, but
not a relevant one I think. Previously the container manager would place
these nodes in /run/systemd/inaccessible/ and that's where PID 1 in the
container would try to add them too when missing. Container manager and
PID 1 in the container would thus manage the same dir together.
With this change the container manager now passes an immutable directory
to the container and leaves /run/systemd entirely untouched, and managed
exclusively by PID 1 inside the container, which is nice to have clear
separation on who manages what.
In order to make sure systemd then usses the /run/host/inaccesible/
nodes this commit changes PID 1 to look for that dir and if it exists
will symlink it to /run/systemd/inaccessible.
Now, this will work fine if new nspawn and new pid 1 in the container
work together. as then the symlink is created and the difference between
the two dirs won't matter.
For the case where an old nspawn invokes a new PID 1: in this case
things work as they always worked: the dir is managed together.
For the case where different container manager invokes a new PID 1: in
this case the nodes aren't typically passed in, and PID 1 in the
container will try to create them and will likely fail partially (though
gracefully) when trying to create char/block device nodes. THis is fine
though as there are fallbacks in place for that case.
For the case where a new nspawn invokes an old PID1: this is were the
(minor) incompatibily happens: in this case new nspawn will place the
nodes in the /run/host/inaccessible/ subdir, but the PID 1 in the
container won't look for them there. Since the nodes are also not
pre-created in /run/systed/inaccessible/ PID 1 will try to create them
there as if a different container manager sets them up. This is of
course not sexy, but is not a total loss, since as mentioned fallbacks
are in place anyway. Hence I think it's OK to accept this minor
incompatibility.
Apparently both Fedora and suse default to btrfs now, it should hence be
good enough for us too.
This enables a bunch of really nice things for us, most importanly we
can resize home directories freely (i.e. both grow *and* shrink) while
online. It also allows us to add nice subvolume based home directory
snapshotting later on.
Also, whenever we mention the three supported types, alaways mention
them in alphabetical order, which is also our new order of preference.
Some masks shouldn't be needed externally, so keep their functions in
the module (others would fit there too but they're used in tests) to
think twice if something would depend on them.
Drop unused function cg_attach_many_everywhere.
Use cgroup_realized instead of cgroup_path when we actually ask for
realized.
This should not cause any functional changes.
When we are about to derealize a controller on v1 cgroup, we first
attempt to delete the controller cgroup and migrate afterwards. This
doesn't work in practice because populated cgroup cannot be deleted.
Furthermore, we leave out slices from migration completely, so
(un)setting a control value on them won't realize their controller
cgroup.
Rework actual realization, unit_create_cgroup() becomes
unit_update_cgroup() and make sure that controller hierarchies are
reduced when given controller cgroup ceased to be needed.
Note that with this we introduce slight deviation between v1 and v2 code
-- when a descendant unit turns off a delegated controller, we attempt
to disable it in ancestor slices. On v2 this may fail (kernel enforced,
because of child cgroups using the controller), on v1 we'll migrate
whole subtree and trim the subhierachy. (Previously, we wouldn't take
away delegated controller, however, derealization was broken anyway.)
Fixes: #14149
