1 |
On Sun, May 15, 2011 at 12:25:32AM +0200, Sven Vermeulen wrote: |
2 |
> I just pushed selinux-base-policy-2.20101213-r15 to hardened-dev.git |
3 |
> overlay. It does not resolve all problems, but at least Gentoo Hardened with |
4 |
> SELinux now boots up properly with OpenRC (and the Gentoo SELinux handbook |
5 |
> has been updated with that what Chris R. said). |
6 |
|
7 |
Small update. I'm going to push out -r16 after the regression tests finish. |
8 |
|
9 |
> But there is still some work ahead. |
10 |
> - rc-update currently *does* *not* *work*. Not good. I know. |
11 |
> The problem is that rc-update (bin_t) calls /sbin/rc (initrc_exec_t) so |
12 |
> transitions to run_init_t which does not have the rights to write in |
13 |
> /etc/runlevels (etc_t). Calling rc-update with run_init doesn't help |
14 |
> either (transitions to initrc_t which also has no rights to write to |
15 |
> etc_t) |
16 |
|
17 |
This is fixed; from -r16, my proposal would be to use an intermediate domain |
18 |
(sysadm_initrc_notrans_t) which, when executing an initrc_exec_t file (like |
19 |
/sbin/rc) transitions back to sysadm_t. |
20 |
|
21 |
The intermediate domain can be entered through an initrc_notrans_exec_t |
22 |
file. |
23 |
|
24 |
> - rc-status works if you use "run_init rc-status". Allowing rc-status to |
25 |
> work without run_init is possible as well (-r15 offers the |
26 |
> gentoo_init_manage_script_status_files interface for this which we can |
27 |
> apply to run_init_t, but you'll also need to add in a |
28 |
> term_use_unallocated_ttys(run_init_t)) but I left it out as I find it to |
29 |
> be an ugly situation then |
30 |
|
31 |
This is fixed as well using the same method. |
32 |
|
33 |
When installing -r16, you want to relabel the /sbin/rc-* and /bin/rc-* files |
34 |
to make use of this though. |
35 |
|
36 |
Wkr, |
37 |
Sven Vermeulen |