Discuss

discuss@lists.centos.org

34758 discussions

usermod and useradd errors
by dnk 22 Apr '07

22 Apr '07

I am not sure if anyone else has run into this, but i did today when setting upo a samba server. http://lists.gobolinux.org/pipermail/gobolinux-users/2007-February/004413.h… In an nutshell you get an error like: "usermod: invalid numeric argument", or the user groups are never assigned, etc. So I have been managing manually. I get the exact same symptoms, etc (but doing things like appending groups, etc). Apparently it can be fixed by downgrading Shadow package to 4.0.17, however with a quick little "rpm -qa | grep shadow", I can see the version with Centos 5 is shadow-utils-4.0.17-12.el5. I have not spent much time digging in deeper as of yet, but wanted ot see if any others have had this issue in Centos 5. DK

2 1

Dumb question (KDE Trash Can)
by Mark Weaver 21 Apr '07

21 Apr '07

Hi All, I've got a really dumb question. I seem to have lost my Trash Can on my KDE desktop and try and as I might I can't get it back. Anyone have any idea how I can create this again or get it back? It's on the desktop as it should be in Gnome, but has disappeared from the KDE desktop. -- Mark "If you have found a very wise man, then you've found a man that at one time was an idiot and lived long enough to learn from his own stupidity." ============================================== Powered by CentOS5 (RHEL5)

4 5

Cron error
by Michael Velez 21 Apr '07

21 Apr '07

Hello, This morning, root received an e-mail from the cron daemon as the following: -------------------------------- Subject: Cron <root> run-parts /etc/cron.daily /etc/cron.daily:00webalizer: Warning: Truncating oversized referrer field -------------------------------- I haven't changed anything to cron or to my intranet website (I'm assuming that's what the webalizer is about) in the last few days. I did do a reboot yesterday which was the first one in a few weeks. Does anybody know what this message is? Michael

3 4

Re: [CentOS] Centos5 compatibilities
by Jay Lee 21 Apr '07

21 Apr '07

Rodrigo Barbosa wrote: > On Tue, Apr 17, 2007 at 05:17:00PM -0700, centos(a)911networks.com wrote: > >> When I am looking for a package that does not exist under rhel5, >> is it better to get rhel4 or fc6? >> > Get RHEL5 and install the fc6 package on it. > Or better yet, grab the FC6 SRPM and recompile it on your RHEL5/FC6 to be sure all your shared libraries, etc match. Very simple to setup and perform as a non-root user: #configure your non-root user's rpm build environment mkdir $HOME/rpm mkdir $HOME/rpm/SOURCES mkdir $HOME/rpm/SPECS mkdir $HOME/rpm/BUILD mkdir $HOME/rpm/SRPMS mkdir $HOME/rpm/RPMS mkdir $HOME/rpm/RPMS/i386 echo "%_topdir $HOME/rpm" >> $HOME/.rpmmacros #build the rpm against your installed platform rpmbuild --rebuild your-fc6.src.rpm Jay

6 10

Learning SELINUX management, help?
by Ben Russo 21 Apr '07

21 Apr '07

OK, so setup CENTOS-5 on a laptop to learn about Xen stuff. KDE Desktop, wanted to print the virt.108.com xen howto. Needed to setup printer first. Open KDE control center, go to printers. Hear error sound, message says "Unable to retrieve the printer list.... Connection to CUPS server failed. ..." So I check to see that cups is running (it is). I check /var/log/messages and see: Apr 19 15:19:52 fox setroubleshoot: SELinux is preventing /usr/sbin/cupsd (cupsd_t) "read write" access to socket:[15083] (rpm_var_lib_t). For complete SELinux messages. run sealert -l 23689411-4e18-4009-b105-779c58b0503f So, I run sealert (as indicated) and see the message about cups (see below for output) I checked in /usr/share/docs/selinux-policy-2.4.6/html and find no references (using grub) for "cupsd_disable_trans" I google on "cupsd_disable_trans" and find no references either. How do I find out what this boolean object is or does? Is there a description of it somewhere? Is it dangerous to just run the command that sealert tells me to run? [root@fox log]# sealert -l 23689411-4e18-4009-b105-779c58b0503f Summary SELinux is preventing /usr/sbin/cupsd (cupsd_t) "read write" access to socket:[15083] (rpm_var_lib_t). Detailed Description SELinux denied access requested by /usr/sbin/cupsd. It is not expected that this access is required by /usr/sbin/cupsd and this access may signal an intrusion attempt. It is also possible that the specific version or configuration of the application is causing it to require additional access. Please file a http://bugzilla.redhat.com/bugzilla/enter_bug.cgi against this package. Allowing Access Sometimes labeling problems can cause SELinux denials. You could try to restore the default system file context for socket:[15083], restorecon -v socket:[15083]. There is currently no automatic way to allow this access. Instead, you can generate a local policy module to allow this access - see http://fedora.redhat.com/docs/selinux-faq-fc5/#id2961385 - or you can disable SELinux protection entirely for the application. Disabling SELinux protection is not recommended. Please file a http://bugzilla.redhat.com/bugzilla/enter_bug.cgi against this package. Changing the "cupsd_disable_trans" boolean to true will disable SELinux protection this application: "setsebool -P cupsd_disable_trans=1." The following command will allow this access: setsebool -P cupsd_disable_trans=1 Additional Information Source Context system_u:system_r:cupsd_t:SystemLow-SystemHigh Target Context system_u:object_r:rpm_var_lib_t Target Objects socket:[15083] [ file ] Affected RPM Packages cups-1.2.4-11.5.1.el5 [application] Policy RPM selinux-policy-2.4.6-30.el5 Selinux Enabled True Policy Type targeted MLS Enabled True Enforcing Mode Enforcing Plugin Name plugins.disable_trans Host Name fox Platform Linux fox 2.6.18-8.el5xen #1 SMP Thu Mar 15 21:02:53 EDT 2007 i686 i686 Alert Count 1 Line Numbers Raw Audit Messages avc: denied { read, write } for comm="cupsd" dev=dm-0 egid=0 euid=0 exe="/usr/sbin/cupsd" exit=0 fsgid=0 fsuid=0 gid=0 items=0 name="__db.000" path="socket:[15083]" pid=5515 scontext=system_u:system_r:cupsd_t:s0-s0:c0.c1023 sgid=0 subj=system_u:system_r:cupsd_t:s0-s0:c0.c1023 suid=0 tclass=file tcontext=system_u:object_r:rpm_var_lib_t:s0 tty=tty1 uid=0 I checked in /usr/share/docs/selinux-policy-2.4.6/html and find no references (using grub) for "cupsd_disable_trans" How do I find out what this boolean object is or does? Is there a description of it somewhere?

4 3

centso 5 modifying my ifcfg-eth[x] files
by Jerry Geis 21 Apr '07

21 Apr '07

I have a machine with 2 network cards. Installing centos 5 on it. I set up both cards with static IP addresses. activate on boot etc... One is e1000 Intel the other is onboard forcedeth (nvidia). After I reboot the forcedeth ifcfg-eth1 file is changed to DHCP. I use netconfig -d eth1 to set it back up and reboot again and the changes back to DHCP. Only thing to add here is the network card does work, however I get a dmesg output that eth1 has an invalid MAC address. Is that invalid MAC address changing my setup? I dont think it should. granted I'm still looking at finding a way to reset he MAC address or something but I dont think the ifcfg-eth files should be modified. This seems like a bug- that is the reason for the post. Great work with centos 5! Jerry

5 5

using command line to setup auto login
by Jerry Geis 21 Apr '07

21 Apr '07

I am trying to find a way to set auto login for a machine under X using the command line. I presume I want to use gconftool-2 to do that. Under the old X centos 4 I edited /etc/X11/gdm/gdm.conf This file no longer exists under X 7 for centos 5. I cant find a way to "show all" under gconftool-2 so that I can see what the key/value pair should be. When I do "gconftool-2 -a " it says it needs a directory. If you dont know what the directory is how can you give one?? I'm confused. I can goto System -> Administration -> Login Screen, then select Security and enable login. However, I need to do this action from the command line. THanks for any pointers that I am missing. Jerry

2 1

SELinux and daemons - clever way to change default locations?
by Florin Andrei 21 Apr '07

21 Apr '07

I'm sure you've seen this before: You need to slightly tweak the default installation of a major daemon - let's say you're running a big MySQL database and you need to put it on a different filesystem, mounted (for example) as /db So you move /var/lib/mysql to /db/mysql (and preserve all the file attributes, including SELinux), change /etc/my.cnf accordingly, start mysqld - and it doesn't work. It turns out you need to tweak SELinux - test the daemon, run audit2allow on the audit log, tweak the policy, test again, repeat until it works. I did this many times, but it strikes me as an inefficient process. Sure, you only do it once per install, but still. I wish there was a simple way to tell SELinux "I moved the MySQL datadir (or the Squid cache dir, or the Cyrus-IMAPd spool) to this new location, but everything else stays the same, please stop bugging me." Any ideas? -- Florin Andrei http://florin.myip.org/

1 2

RE: seamonkey?
by Russel 21 Apr '07

21 Apr '07

What happened to seamonkey? I thought firefox was retired in 4 and replaced by seamonkey. Imagine my surprise when 5 came with firefox. What's a guy to do? TIA Russel

5 5

can't find python24 rpms for centos4 64bit anymore!
by sophana 21 Apr '07

21 Apr '07

Hi I'm using python24 under centos4 for a webware application. I downloaded it from atrpms repository some time ago. I'm installing a new VPS with centos4 64bit, and can't find any python24 package! python.org pyvault doesn't support 64 bit architectures, and atrpms don't make python24 packages anymore. I don't have 64 bit architecture at home so I can't recompile python24 on my (small) vps. Please help! Regards

1 0

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Discuss