<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN"> <HTML> <HEAD> <META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8"> <META NAME="GENERATOR" CONTENT="GtkHTML/3.0.10"> </HEAD> <BODY> Yeh.. I know, I should give it a better look but because I messed up very badly om my actual work box I have to make-do with this testbox. read: I ain't touching anything :) I'll tinker with selinux later on, but I'm pretty sure I didn't kill it though. On Sun, 2004-04-11 at 00:19, Hannes Mayer wrote: <BLOCKQUOTE TYPE=CITE> <PRE>Yves Vanlerberghe wrote: > I did a yum update yesterday and the whole box just gave up at the point > when the filesystem gets mounted. It said that only root could mount it > and then started spewing out some SELinux errors. Dunno whether this > will help but I got it sorted after adding 'selinux=0' as argument to > the kernel. There's a great faq on the topic somewhere on the redhat site. > I think something got screwed up in the policy file so I'll wait for an > updated policy to come trough via Yum before I give selinux another test.. I also just had a strange experience with SELinux: I disabled it in /etc/sysconfig/selinux (SELINUX=disabled) in order to run "updatedb", then reenabled it and after that the screen was flooded by audit klogd messages. I logged in via ssh from another machine and did a "fixfiles relabel" That will set the SELinux permissions on all files and after that I did't have any audit-klogd messages. But sorry, I'm somewhat clueless if it already fails at /sbin/init... Maybe via an emergency CD ? Cheers, Hannes. PS: to disable SELinux temporarily (i.e. to run a command that won't with SELinux enabled) use "setenforce 0" and "setenforce 1" to switch it on again. </PRE> </BLOCKQUOTE> </BODY> </HTML>