Red Hat Bugzilla – Full Text Bug Listing
|Summary:||Using a non-SELinux kernel possibly causing id spewage?|
|Product:||[Fedora] Fedora||Reporter:||Tim Powers <timp>|
|Component:||coreutils||Assignee:||Daniel Walsh <dwalsh>|
|Status:||CLOSED RAWHIDE||QA Contact:|
|Fixed In Version:||Doc Type:||Bug Fix|
|Doc Text:||Story Points:||---|
|Last Closed:||2004-02-11 09:19:02 EST||Type:||---|
|oVirt Team:||---||RHEL 7.3 requirements from Atomic Host:|
Description Tim Powers 2004-01-21 13:47:54 EST
We created a chroot environment to run mass rebuild tests for the current devel stream for what will eventually become Test 1 for FC2. The main OS installed outside of the chroot is RHEL 3 U1. When I chroot into the chrooted environment I see these messages: [root@loki rebuildtest]# /usr/sbin/chroot . id: can't get process context id: can't get process context id: can't get process context bash: [: too many arguments [root@loki /]# Now, this may be /proc not being mounted or something else odd, I'm not sure. Elliot said there's a bug WRT SELinux and coreutils, so I'm filing this through his recommendation and setting the owner to dwalsh.
Comment 1 Daniel Walsh 2004-01-21 14:28:29 EST
This is caused by a change in the definition of is_selinux_enabled. This used to be a boolean function, but now it returns >0 if enabled 0 if not enabled and -1 if unknown. IE /proc not mounted. Most of the SELinux code is written expecting the boolean. I have rewritten the coreutils patch to fix this problem. I need to look at other kits to see what the correct thing to do is. Dan
Comment 2 Daniel Walsh 2004-02-11 09:19:02 EST
Fixed in rawhide.