similar to: deliver triggering SELinux AVC denials

I'm trying to set the context on an nfs mounted /home. I believe exactly like in Redhat's Deployment Guide at http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/5.2/html/Deployment_Guide/ch45s02s03.html On my system running CentOS 5.2: $ ls -alZ /home drwxr-xr-x root root system_u:object_r:home_root_t . drwxr-xr-x root root system_u:object_r:root_t .. $ mount -t

Not a problem ... sharing a solution (this time)! Please correct my understanding of the process, if required. "i_stream_read() failed: Permission denied" is an error message generated when a large-ish file (>128kb in my case) is attached to a message that has been passed to Dovecot's deliver program when SELinux is being enforced. In my case, these messages are first run

Hi all, I just installed a CentOS 5 machine from Kickstart. I configure NSS and PAM to lookup and authenticate users from LDAP with authconfig. On my LDAP I also have some automount configuration, but I'm not running automount on this server. SELinux is installed and enforcing. Whenever I try to install an RPM (and in other occasions during boot) I see those messages: # rpm -Uvh ... .rpm

Hello, I was using CentOS 5.5 as a "playground" VM at my WinXP notebook and now I'm migrating to a new CentOS 5.6 install and everything has worked well - except samba. I have this very permissive config to export my ~/src dir: # cat /etc/samba/smb.conf [global] guest ok = yes guest account = afarber security = share hosts allow = 172.16.6. 127.0.0.1 [src]

Hi, I'm running dovecot (1.1.7) deliver and sieve (1.1.5) on a Fedora 9 platform, using selinux targetet mode. Most of the mail deliveries goes well, but once deliver tried to copy the mail to the /tmp directory, which it seems it not allowed by selinux. I guess that deliver wants to sanitize the mail or something and therefore copies it to /tmp. Before I ask for selinux to allow this, I

Hi All, I'm running CentOS 5.2 with SELinux in enforcing mode (default targeted policy). The server hosts a PHP web app that sends mail. I'm getting the following errors (see end of message) in my selinux audit.log file every time the app sends an email. The email always seems to get sent successfully, despite the log messages. However, they do concern me and I would like to understand

Hi. I've installed BackupPC 3.1.0 from Testing repository, to Cent OS 5.2 x86_64, and I am hitting an SE Linux denial - the httpd cannot talk to the BackupPC socket: type=AVC msg=audit(07/31/2008 17:18:53.623:410) : avc: denied { connectto } for pid=11767 comm=httpd path=/var/log/BackupPC/BackupPC.sock scontext=user_u:system_r:httpd_t:s0 tcontext=user_u:system_r:initrc_t:s0

Hi all, On my newly up-and-running nameserver (CentOS 5), I noticed the following alerts in /var/log/messages after restarting BIND. (lines inserted to aid in reading). As I'm new to SELinux, I'm hoping for some pointers on 1) if this is an issue which simply *must* be addressed, or if it's something I should live with, and 2) how to eliminate the warming messages without sacrificing

Any ideas why bind is putting the tmp files in the [chroot]/var/named directory and not in /tmp or /var/tmp? [root at devserver21 chroot]# Aug 15 14:08:21 devserver21 named[5101]: loading configuration from '/etc/named.conf' Aug 15 14:08:21 devserver21 named: named reload succeeded Aug 15 14:08:21 devserver21 named[5101]: dumping master file: tmp-XXXXQ5X9mC: open: permission denied Aug 15

Hi: I have been using Dovecot for well over a year now and it has always worked with few problems. The mail setup is not simple... Postfix+MailScanner+ClamAV+Docvecot+MySql+postfix.admin... just to mention the major things. The system is CentOS 5.3 on VMware. The maildir is on an NFS share, index and control is local. About a month ago I thought I upgraded from 1.1.x to 1.2.x. by doing an

Hi. I'm trying to setup squid with SELinux, the problem i encounter is taht i want to add another directory for cache, in this system we have a home partition with huge space, i create a squid dir and add the path with semanage: semanage fcontext -a -t squid_cache_t '/home/squid(/.*)?' i check the files and are in the good context: drwxr-xr-x squid squid

Yesterday I activated SELinux in targeted mode, then I rebooted and started receiving some error messages in the system services initialization: ====================================================================== audit(1156518721.252:2): avc: denied { read } for pid=2223 comm="syslogd" name="libc-2.3.4.so" dev=dm-0 ino=50441 scontext=user_u:system_r:syslogd_t

I have some centos 4.4 server. i have disable selinux for some software problem: # cat /etc/selinux/config # This file controls the state of SELinux on the system. # SELINUX= can take one of these three values: # enforcing - SELinux security policy is enforced. # permissive - SELinux prints warnings instead of enforcing. # disabled - SELinux is fully disabled. SELINUX=disable #

Hi All, I have this following issue in SELinux. I did what instruction said but the security context has still never changed. Do I need to create local SELinux module? I hope anyone could help me out of this. Thank you. ------------------------------------------------------- # sealert -b ........................................ Summary: SELinux is preventing postmaster (postgresql_t)

CentOS 4 - updated to current, rebooted to new kernel and now I can't get mysqld to start... # service mysqld start Timeout error occurred trying to start MySQL Daemon #tail -n 4 /var/log/messages Nov 12 00:48:56 srv1 kernel: audit(1131781736.221:4): avc: denied { write } for pid=4874 comm="mysqld" name="tmp" dev=dm-0 ino=2894305 scontext=root:system_r:mysqld_t

Hi, I am not very experienced with SELinux and I have a problem which I can't track down. Any help would be really appreciated. I have an 'install everything' Centos 4.2 system which I am using as a workstation. Before anyone tells me off for installing everything, I have done this in order to get used to CentOS before using it on live servers. Anyway when I log into X (gnome, gdm)

Hey all... not exactly an SELinux veteran, but am trying to work through some issues. Specifically, setting up a simple Samba configuration on a CentOS 5 machine. Determined I needed to do setsebool -P samba_enable_home_dirs 1 In order to get access to home directory shares working correctly. Fine; this is documented in samba_selinux(8). However, I still see the following in my

Hello, Running httpd-2.2.3-43.el5.centos.3 on CentOS release 5.5 (Final), I have : $ ps -Ze LABEL PID TTY TIME CMD user_u:system_r:httpd_t 12833 ? 00:00:00 httpd Is it normal for httpd to have this context (user_u:system_r:httpd_t) ? I was expecting system_u:system_r:httpd_t. And if it is not normal, is it because I have restarted httpd by

Hello, I received the below SELinux message today and I am trying to figure out what caused it. I see what it says under Allow Access but I am not sure this is what I really want to do without know why it happened in the first place. What should I be looking at to understand what or why this has happened? Any help I would be most grateful for. Here is the output form SELinux SUMMARY:

I am configuring a new mail server on RHEL 5 x64. I have configured dovecot as follows: ... protocols = imaps ... ssl_cert_file = /etc/pki/dovecot/certs/mailserver.cer ssl_key_file = /etc/pki/dovecot/private/mailserver.pem ... login_process_size = 64 ... mail_location = maildir:~/Maildir ... passdb pam { args = "session=yes cache_key=%u%s dovecot" } ... I'm getting the following