RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.
Bug 743222 - Cannot run screen in MLS mode
Summary: Cannot run screen in MLS mode
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Enterprise Linux 6
Classification: Red Hat
Component: initscripts
Version: 6.2
Hardware: All
OS: Linux
urgent
urgent
Target Milestone: rc
: ---
Assignee: initscripts Maintenance Team
QA Contact: Miroslav Vadkerti
URL:
Whiteboard: Regression
Depends On:
Blocks: RHEL62CCC 739953 747120 846801 846802
TreeView+ depends on / blocked
 
Reported: 2011-10-04 09:52 UTC by Miroslav Vadkerti
Modified: 2012-08-08 18:29 UTC (History)
4 users (show)

Fixed In Version: initscripts-9.03.27-1.el6
Doc Type: Bug Fix
Doc Text:
Cause restorecon utility changed its behaviour so it doesn't change MLS level unless -F parameter is used. Consequence /dev and /dev/pts filesystems were not correctly labelled after boot in systems with configured MLS policy Fix restorecon -F is now used for /dev and /dev/pts
Clone Of:
Environment:
Last Closed: 2011-12-06 18:10:09 UTC
Target Upstream Version:
Embargoed:

Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2011:1528 0 normal SHIPPED_LIVE initscripts bug fix and enhancement update 2011-12-06 01:02:05 UTC

Description Miroslav Vadkerti 2011-10-04 09:52:06 UTC 
Description of problem:
When trying to login via screen as staff_u in MLS I get a AVC denial when screen is trying to access PTY

type=SYSCALL msg=audit(1317717208.934:77): arch=c000003e syscall=59 success=yes exit=0 a0=1503480 a1=1503430 a2=150b8b0 a3=7fffe1123280 items=0 ppid=1685 pid=1711 auid=500 uid=500 gid=500 euid=500 suid=500 fsuid=500 egid=500 sgid=500 fsgid=500 tty=tty2 ses=3 comm="screen" exe="/usr/bin/screen" subj=staff_u:staff_r:staff_screen_t:s0-s15:c0.c1023 key=(null)
type=AVC msg=audit(1317717208.981:78): avc:  denied  { search } for  pid=1738 comm="screen" name="/" dev=devpts ino=1 scontext=staff_u:staff_r:staff_screen_t:s0-s15:c0.c1023 tcontext=system_u:object_r:devpts_t:s15:c0.c1023 tclass=dir

Version-Release number of selected component (if applicable):
selinux-policy-3.7.19-113.el6

How reproducible:
100%

Steps to Reproduce:
1. login as staff_u user
2. run screen
  
Actual results:
screen fails saying no PTY found

Expected results:
screen works OK
Comment 2 Miroslav Grepl 2011-10-04 11:46:11 UTC 
Mirek,
how is labeled /dev?
Comment 3 Miroslav Vadkerti 2011-10-04 13:00:38 UTC 
Looks like this is a bug in initscripts because of the restorecon -F fix in policycoreutils:

mount -n /dev/pts >/dev/null 2>&1
[ -n "$SELINUX_STATE" ] && restorecon /dev/pts >/dev/null 2>&1

should be


mount -n /dev/pts >/dev/null 2>&1
[ -n "$SELINUX_STATE" ] && restorecon -F /dev/pts >/dev/null 2>&1
Comment 5 Miroslav Grepl 2011-10-04 13:09:42 UTC 
well, not only for "restorecon -F /dev/pts" since I see /dev mislabled after booting.

I needed also change

-    /sbin/restorecon -R /dev 2>/dev/null
+    /sbin/restorecon -R -F /dev 2>/dev/null
Comment 6 Daniel Walsh 2011-10-04 14:00:24 UTC 
What is getting mislabeled?  -F should not be required to fix the types and the levels should all be s0?
Comment 7 Daniel Walsh 2011-10-04 14:08:20 UTC 
Ok this looks like it was caused by a change in restorecon's behaviour to not change the label of a directory if the only thing different is the Level.  In this case the system seems to be creating directories at SystemHigh on boot and then needs to fix the directories labels, we should use the -F in this case to force the change to the default.
Comment 8 Eduard Benes 2011-10-04 14:48:32 UTC 
(In reply to comment #7)
> Ok this looks like it was caused by a change in restorecon's behaviour to not
> change the label of a directory if the only thing different is the Level.  In
> this case the system seems to be creating directories at SystemHigh on boot and
> then needs to fix the directories labels, we should use the -F in this case to
> force the change to the default.

The problem is, that all the scripts expecting the "old default" behaviour will be broken with the new restorecon. Could we keep the "old default" behaviour and for example add a new option to satisfy the new requirements which led us to the change in restorecon?
Comment 9 Bill Nottingham 2011-10-04 17:51:28 UTC 
Yes, I'd be concerned that we will break more than just initscripts. (Modulo the number of people that actually use MLS mode, of course.)
Comment 10 Daniel Walsh 2011-10-04 20:41:22 UTC 
Well the fix was added for MLS mode, since there is not very often when the we see an MLS/MCS label outside of s0.

The fix will also prevent people from accidently changing the labels on sandboxes or svirt processes.
Comment 12 Bill Nottingham 2011-10-05 16:45:06 UTC 
Dan - how many potential situations are there like this where restorecon invocation would need to be changed?
Comment 18 Petr Lautrbach 2011-10-07 14:06:20 UTC 
http://git.fedorahosted.org/git/?p=initscripts.git;a=commitdiff;h=c21d4c974b2305c8ce6d429888be61d886fdcd44
Comment 21 Petr Lautrbach 2011-10-14 12:59:56 UTC 
    Technical note added. If any revisions are required, please edit the "Technical Notes" field
    accordingly. All revisions will be proofread by the Engineering Content Services team.
    
    New Contents:
Cause
    restorecon utility changed its behaviour so it doesn't change MLS level 
unless -F parameter is used.

Consequence
    /dev and /dev/pts filesystems were not correctly labelled after boot in 
systems with configured MLS policy

Fix
    restorecon -F is now used for /dev and /dev/pts
Comment 22 errata-xmlrpc 2011-12-06 18:10:09 UTC 
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-2011-1528.html
Note You need to log in before you can comment on or make changes to this bug.