Add a first outline of the integration points of osbuild and
selinux, from two perspectives: how osbuild interacts with
the host and how osbuild labels the target artefact and the
resulting peculiarties.
Allow osbuild_t to no_new_privs (nnp) and nosuid domain transition
into setfiles_mac_t and install_t. nnp is a inheritable per-thread
flag (PR_SET_NO_NEW_PRIVS, see prctl(2)), whereby a promise is made
by execve(2) to not grant any new privileges that could not have
been done without the execv call. This is on contrast to what can
be done via SELinux rules, i.e. in our case `setfiles_mac_t` and
`install_t` can set arbitrary SELinux labels, but `osbuild_t`
itself can not; but `osbuild_t` enables the transitioning of
`setfiles_mac_t` for the `setfiles` binary via execve(2) from a
process with `osbuild_t`. Related, the nosuid mount flag, prevents
the suid, sgid bits to be interpreted and thus are in the same
spirit as nnp, i.e. no new privs during execve(2).
Thus SELinux domain transitions stand in contrast with nnp and
nosuid transitions, and have therefore been de-coupled. See also
the corresponding kernel patch at [1] for more information.
bubblewrap (bwrap) in contrast to `systemd-nspawn` always sets the
nnp flag, as well as the nosuid option for all bind-mounts. Since
we no use bwrap to contain processes we need to allow the nnp and
nosuid transitions from `osbuild_t` to `setfiles_mac_t` and
`install_t`.
[1] https://patchwork.kernel.org/patch/9841441/
When osbuild is invoked via the osbuild composer worker, the latter
is run in the unconfined_service_t domain, so add a rule that
allows that transition.
A usual step in creating OS file system trees is to apply the
correct SELinux labels for all files and directories. This is
done by the org.osbuild.selinux stage, which internally uses the
setfiles command in order to do so. The SELiunx policy to be
used for this operation is the one of the newly created system,
not the host one. It therefore can contain labels that are not
known on the host. The kernel will prevent setting invalid,
i.e. unknown, labels unless the caller has the CAP_MAC_ADMIN
capability. By default, setfiles is executed in the setfiles_t
domain, where it lacks that capability. Therefore a custom
osbuild SELinux policy was created, with a special transition
rule that will execute setfiles in the setfiles_mac_t
domain. All stages, sources and assemblers as well as the main
binary are label with the new osbuild_exec_t label.
Additionally, allow a transition from osbuild_t to install_t by
using `anaconda_domtrans_install`, so that ostree and
rpm-ostree, which are labeled as install_exec_t, can transition
to the install_t domain when called form osbuild. Update the
spec file to build the policy and include it in a new
osbuild-selinux sub-package.
