Bug 736153
Summary: | restorecon should not change USER,ROLE,MLS range unless -F is specified | ||
---|---|---|---|
Product: | Red Hat Enterprise Linux 6 | Reporter: | Daniel Walsh <dwalsh> |
Component: | policycoreutils | Assignee: | Daniel Walsh <dwalsh> |
Status: | CLOSED ERRATA | QA Contact: | Milos Malik <mmalik> |
Severity: | urgent | Docs Contact: | |
Priority: | urgent | ||
Version: | 6.1 | CC: | dwalsh, ebenes, ksrot, mgrepl, mmalik, mvadkert, psklenar, sgrubb |
Target Milestone: | rc | ||
Target Release: | 6.2 | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | policycoreutils-2.0.83-19.14.el6 | Doc Type: | Bug Fix |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2011-12-06 15:31:08 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: | |||
Bug Depends On: | |||
Bug Blocks: | 584498, 743047, 846801, 846802 |
Description
Daniel Walsh
2011-09-06 20:29:35 UTC
Fixed in selinux-policy- 2.0.83-19.14 I'm correcting "Fixed In Version" field. More info in Changelog at https://brewweb.devel.redhat.com/buildinfo?buildID=179639. Yes This bug is causing regressions (currently 2) but I really don't like this behaviour change: See please: https://bugzilla.redhat.com/show_bug.cgi?id=741430 https://bugzilla.redhat.com/show_bug.cgi?id=743222 Dan why do we need this fix. Is some customer behind this? Thanks! Yes this fix was for MLS machines and soon will be needed for Targetd/Containers. The idea is we want restorecon to fix the "type" portion of the label but not the level or the user, unless -F is given. The current behaviour does not allow us to label a users homedirectory correctly if they are running at a level higher then s0. I believe this bug should be set back to ASSIGNED and we should discuss it before including into RHEL 6.2. I have a couple of questions on my mind: Does the change just move a part of previous default functionality to -F option? Is there a new functionality added to the -F option? Could we keep the previous functionality and add a new option to take care of the requirement for labeling homedirs you've mentioned in previous comment? The short summary is, that the new version of restorecon changes default behaviour when used without -F option. With the new version one has to use -F to get the same results as before. This is causing regressions as mentioned in the previous comment in system scripts counting on the previous behaviour (not mentioning possible issues in customer scripts). OLD behaviour withOUT -F: $ chcon -u staff_u -t shadow_t -l s0:c1 test ; ls -Z test ; restorecon -vv test ; ls -Z test -rw-rw-r--. estragon estragon staff_u:object_r:shadow_t:s0:c1 test restorecon reset /home/estragon/test context staff_u:object_r:shadow_t:s0:c1->unconfined_u:object_r:user_home_t:s0 -rw-rw-r--. estragon estragon unconfined_u:object_r:user_home_t:s0 test NEW behaviour withOUT -F: # chcon -u staff_u -t shadow_t -l s0:c1 test ; ls -Z test ; restorecon -vv test ; ls -Z test -rw-r--r--. root root staff_u:object_r:shadow_t:s0:c1 test restorecon reset /root/test context staff_u:object_r:shadow_t:s0:c1->staff_u:object_r:admin_home_t:s0:c1 -rw-r--r--. root root staff_u:object_r:admin_home_t:s0:c1 test ---- OLD behaviour WITH -F: $ chcon -u staff_u -t shadow_t -l s0:c1 test ; ls -Z test ; restorecon -Fvv test ; ls -Z test -rw-rw-r--. estragon estragon staff_u:object_r:shadow_t:s0:c1 test restorecon reset /home/estragon/test context staff_u:object_r:shadow_t:s0:c1->unconfined_u:object_r:user_home_t:s0 -rw-rw-r--. estragon estragon unconfined_u:object_r:user_home_t:s0 test NEW behaviour WITH -F: # chcon -u staff_u -t shadow_t -l s0:c1 test ; ls -Z test ; restorecon -Fvv test ; ls -Z test -rw-r--r--. root root staff_u:object_r:shadow_t:s0:c1 test restorecon reset /root/test context staff_u:object_r:shadow_t:s0:c1->system_u:object_r:admin_home_t:s0 -rw-r--r--. root root system_u:object_r:admin_home_t:s0 test Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2011-1637.html |