Table of Contents
The Linux kernel has a built-in stateful firewall named iptables. To stop the iptables firewall on Red Hat, use the service command.
root@RHELv4u4:~# service iptables stop Flushing firewall rules: [ OK ] Setting chains to policy ACCEPT: filter [ OK ] Unloading iptables modules: [ OK ] root@RHELv4u4:~#
The easy way to configure iptables, is to use a graphical tool like KDE's kmyfirewall or Security Level Configuration Tool. You can find the latter in the graphical menu, somewhere in System Tools - Security, or you can start it by typing system-config-securitylevel in bash. These tools allow for some basic firewall configuration. You can decide whether to enable or disable the firewall, and what typical standard ports are allowed when the firewall is active. You can even add some custom ports. When you are done, the configuration is written to /etc/sysconfig/iptables on Red Hat.
root@RHELv4u4:~# cat /etc/sysconfig/iptables # Firewall configuration written by system-config-securitylevel # Manual customization of this file is not recommended. *filter :INPUT ACCEPT [0:0] :FORWARD ACCEPT [0:0] :OUTPUT ACCEPT [0:0] :RH-Firewall-1-INPUT - [0:0] -A INPUT -j RH-Firewall-1-INPUT -A FORWARD -j RH-Firewall-1-INPUT -A RH-Firewall-1-INPUT -i lo -j ACCEPT -A RH-Firewall-1-INPUT -p icmp --icmp-type any -j ACCEPT -A RH-Firewall-1-INPUT -p 50 -j ACCEPT -A RH-Firewall-1-INPUT -p 51 -j ACCEPT -A RH-Firewall-1-INPUT -p udp --dport 5353 -d 224.0.0.251 -j ACCEPT -A RH-Firewall-1-INPUT -p udp -m udp --dport 631 -j ACCEPT -A RH-Firewall-1-INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT -A RH-F...NPUT -m state --state NEW -m tcp -p tcp --dport 22 -j ACCEPT -A RH-F...NPUT -m state --state NEW -m tcp -p tcp --dport 80 -j ACCEPT -A RH-F...NPUT -m state --state NEW -m tcp -p tcp --dport 21 -j ACCEPT -A RH-F...NPUT -m state --state NEW -m tcp -p tcp --dport 25 -j ACCEPT -A RH-Firewall-1-INPUT -j REJECT --reject-with icmp-host-prohibited COMMIT root@RHELv4u4:~#
To start the service, issue the service iptables start command. You can configure iptables to start at boot time with chkconfig.
root@RHELv4u4:~# service iptables start Applying iptables firewall rules: [ OK ] root@RHELv4u4:~# chkconfig iptables on root@RHELv4u4:~#
One of the nice features of iptables is that it displays extensive status information when queried with the service iptables status command.
root@RHELv4u4:~# service iptables status Table: filter Chain INPUT (policy ACCEPT) target prot opt source destination RH-Firewall-1-INPUT all -- 0.0.0.0/0 0.0.0.0/0 Chain FORWARD (policy ACCEPT) target prot opt source destination RH-Firewall-1-INPUT all -- 0.0.0.0/0 0.0.0.0/0 Chain OUTPUT (policy ACCEPT) target prot opt source destination Chain RH-Firewall-1-INPUT (2 references) target prot opt source destination ACCEPT all -- 0.0.0.0/0 0.0.0.0/0 ACCEPT icmp -- 0.0.0.0/0 0.0.0.0/0 icmp type 255 ACCEPT esp -- 0.0.0.0/0 0.0.0.0/0 ACCEPT ah -- 0.0.0.0/0 0.0.0.0/0 ACCEPT udp -- 0.0.0.0/0 224.0.0.251 udp dpt:5353 ACCEPT udp -- 0.0.0.0/0 0.0.0.0/0 udp dpt:631 ACCEPT all -- 0.0.0.0/0 0.0.0.0/0 state RELATED,ESTABLISHED ACCEPT tcp -- 0.0.0.0/0 0.0.0.0/0 state NEW tcp dpt:22 ACCEPT tcp -- 0.0.0.0/0 0.0.0.0/0 state NEW tcp dpt:80 ACCEPT tcp -- 0.0.0.0/0 0.0.0.0/0 state NEW tcp dpt:21 ACCEPT tcp -- 0.0.0.0/0 0.0.0.0/0 state NEW tcp dpt:25 REJECT all -- 0.0.0.0/0 0.0.0.0/0 reject-with icmp-host-prohibited root@RHELv4u4:~#
Mastering firewall configuration requires a decent knowledge of tcp/ip. Good iptables tutorials can be found online here http://iptables-tutorial.frozentux.net/iptables-tutorial.html and here http://tldp.org/HOWTO/IP-Masquerade-HOWTO/.
1. Verify whether the firewall is running.
root@rhel55 ~# service iptables status | head Table: filter Chain INPUT (policy ACCEPT) num target prot opt source destination 1 RH-Firewall-1-INPUT all -- 0.0.0.0/0 0.0.0.0/0 Chain FORWARD (policy ACCEPT) num target prot opt source destination 1 RH-Firewall-1-INPUT all -- 0.0.0.0/0 0.0.0.0/0 Chain OUTPUT (policy ACCEPT)
2. Stop the running firewall.
root@rhel55 ~# service iptables stop Flushing firewall rules: [ OK ] Setting chains to policy ACCEPT: filter [ OK ] Unloading iptables modules: [ OK ] root@rhel55 ~# service iptables status Firewall is stopped.
Back when resources like RAM memory were limited, a super-server was devised to listen to all sockets and start the appropriate daemon only when needed. Services like swat, telnet and ftp are typically served by such a super-server. The xinetd superdaemon is more recent than inetd. We will discuss the configuration both daemons.
Recent Linux distributions like RHEL5 and Ubuntu10.04 do not activate inetd or xinetd by default, unless an application requires it.
First verify whether your computer is running inetd or xinetd. This Debian 4.0 Etch is running inetd.
root@barry:~# ps fax | grep inet 3870 ? Ss 0:00 /usr/sbin/inetd
This Red Hat Enterprise Linux 4 update 4 is running xinetd.
[root@RHEL4b ~]# ps fax | grep inet 3003 ? Ss 0:00 xinetd -stayalive -pidfile /var/run/xinetd.pid
Both daemons have the same functionality (listening to many ports, starting other daemons when they are needed), but they have different configuration files.
The xinetd daemon is often called a superdaemon because it listens to a lot of incoming connections, and starts other daemons when they are needed. When a connection request is received, xinetd will first check TCP wrappers (/etc/hosts.allow and /etc/hosts.deny) and then give control of the connection to the other daemon. This superdaemon is configured through /etc/xinetd.conf and the files in the directory /etc/xinetd.d. Let's first take a look at /etc/xinetd.conf.
paul@RHELv4u2:~$ cat /etc/xinetd.conf
#
# Simple configuration file for xinetd
#
# Some defaults, and include /etc/xinetd.d/
defaults
{
instances = 60
log_type = SYSLOG authpriv
log_on_success = HOST PID
log_on_failure = HOST
cps = 25 30
}
includedir /etc/xinetd.d
paul@RHELv4u2:~$
According to the settings in this file, xinetd can handle 60 client requests at once. It uses the authpriv facility to log the host ip-address and pid of successful daemon spawns. When a service (aka protocol linked to daemon) gets more than 25 cps (connections per second), it holds subsequent requests for 30 seconds.
The directory /etc/xinetd.d contains more specific configuration files. Let's also take a look at one of them.
paul@RHELv4u2:~$ ls /etc/xinetd.d
amanda chargen-udp echo klogin rexec talk
amandaidx cups-lpd echo-udp krb5-telnet rlogin telnet
amidxtape daytime eklogin kshell rsh tftp
auth daytime-udp finger ktalk rsync time
chargen dbskkd-cdb gssftp ntalk swat time-udp
paul@RHELv4u2:~$ cat /etc/xinetd.d/swat
# default: off
# description: SWAT is the Samba Web Admin Tool. Use swat \
# to configure your Samba server. To use SWAT, \
# connect to port 901 with your favorite web browser.
service swat
{
port = 901
socket_type = stream
wait = no
only_from = 127.0.0.1
user = root
server = /usr/sbin/swat
log_on_failure += USERID
disable = yes
}
paul@RHELv4u2:~$
The services should be listed in the /etc/services file. Port determines the service port, and must be the same as the port specified in /etc/services. The socket_type should be set to stream for tcp services (and to dgram for udp). The log_on_failure += concats the userid to the log message formatted in /etc/xinetd.conf. The last setting disable can be set to yes or no. Setting this to no means the service is enabled!
Check the xinetd and xinetd.conf manual pages for many more configuration options.
This superdaemon has only one configuration file /etc/inetd.conf. Every protocol or daemon that it is listening for, gets one line in this file.
root@barry:~# grep ftp /etc/inetd.conf tftp dgram udp wait nobody /usr/sbin/tcpd /usr/sbin/in.tftpd /boot/tftp root@barry:~#
You can disable a service in inetd.conf above by putting a # at the start of that line. Here an example of the disabled vmware web interface (listening on tcp port 902).
paul@laika:~$ grep vmware /etc/inetd.conf #902 stream tcp nowait root /usr/sbin/vmware-authd vmware-authd
1. Verify on all systems whether they are using xinetd or inetd.
2. Look at the configuration files.
3. (If telnet is installable, then replace swat in these questions with telnet) Is swat installed ? If not, then install swat and look at the changes in the (x)inetd configuration. Is swat enabled or disabled ?
4. Disable swat, test it. Enable swat, test it.
The older nfs versions 2 and 3 are stateless (udp) by default, but they can use tcp. Clients connect to the server using rpc (on Linux this is controlled by the portmap daemon. Look at rpcinfo to verify that nfs and its related services are running.
root@RHELv4u2:~# /etc/init.d/portmap status portmap (pid 1920) is running... root@RHELv4u2:~# rpcinfo -p program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100024 1 udp 32768 status 100024 1 tcp 32769 status root@RHELv4u2:~# service nfs start Starting NFS services: [ OK ] Starting NFS quotas: [ OK ] Starting NFS daemon: [ OK ] Starting NFS mountd: [ OK ]
The same rpcinfo command when nfs is started.
root@RHELv4u2:~# rpcinfo -p program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100024 1 udp 32768 status 100024 1 tcp 32769 status 100011 1 udp 985 rquotad 100011 2 udp 985 rquotad 100011 1 tcp 988 rquotad 100011 2 tcp 988 rquotad 100003 2 udp 2049 nfs 100003 3 udp 2049 nfs 100003 4 udp 2049 nfs 100003 2 tcp 2049 nfs 100003 3 tcp 2049 nfs 100003 4 tcp 2049 nfs 100021 1 udp 32770 nlockmgr 100021 3 udp 32770 nlockmgr 100021 4 udp 32770 nlockmgr 100021 1 tcp 32789 nlockmgr 100021 3 tcp 32789 nlockmgr 100021 4 tcp 32789 nlockmgr 100005 1 udp 1004 mountd 100005 1 tcp 1007 mountd 100005 2 udp 1004 mountd 100005 2 tcp 1007 mountd 100005 3 udp 1004 mountd 100005 3 tcp 1007 mountd root@RHELv4u2:~#
nfs version 4 requires tcp (port 2049) and supports Kerberos user authentication as an option. nfs authentication only takes place when mounting the share. nfs versions 2 and 3 authenticate only the host.
nfs is configured in /etc/exports. Here is a sample /etc/exports to explain the syntax. You need some way (NIS domain or LDAP) to synchronize userid's across computers when using nfs a lot. The rootsquash option will change UID 0 to the UID of the nfsnobody user account. The sync option will write writes to disk before completing the client request.
paul@laika:~$ cat /etc/exports # Everyone can read this share /mnt/data/iso *(ro) # Only the computers barry and pasha can readwrite this one /var/www pasha(rw) barry(rw) # same, but without root squashing for barry /var/ftp pasha(rw) barry(rw,no_root_squash) # everyone from the netsec.lan domain gets access /var/backup *.netsec.lan(rw) # ro for one network, rw for the other /var/upload 192.168.1.0/24(ro) 192.168.5.0/24(rw)
You don't need to restart the nfs server to start exporting your newly created exports. You can use the exportfs -va command to do this. It will write the exported directories to /var/lib/nfs/etab, where they are immediately applied.
We have seen the mount command and the /etc/fstab file before.
root@RHELv4u2:~# mount -t nfs barry:/mnt/data/iso /home/project55/ root@RHELv4u2:~# cat /etc/fstab | grep nfs barry:/mnt/data/iso /home/iso nfs defaults 0 0 root@RHELv4u2:~#
Here is another simple example. Suppose the project55 people tell you they only need a couple of CD-ROM images, and you already have them available on an nfs server. You could issue the following command to mount this storage on their /home/project55 mount point.
root@RHELv4u2:~# mount -t nfs 192.168.1.40:/mnt/data/iso /home/project55/ root@RHELv4u2:~# ls -lh /home/project55/ total 3.6G drwxr-xr-x 2 1000 1000 4.0K Jan 16 17:55 RHELv4u1 drwxr-xr-x 2 1000 1000 4.0K Jan 16 14:14 RHELv4u2 drwxr-xr-x 2 1000 1000 4.0K Jan 16 14:54 RHELv4u3 drwxr-xr-x 2 1000 1000 4.0K Jan 16 11:09 RHELv4u4 -rw-r--r-- 1 root root 1.6G Oct 13 15:22 sled10-vmwarews5-vm.zip root@RHELv4u2:~#
1. Create two directories with some files. Use nfs to share one of them as read only, the other must be writable. Have your neighbour connect to them to test.
2. Investigate the user owner of the files created by your neighbour.
3. Protect a share by ip-address or hostname, so only your neighbour can connect.