Raspberry Pi IPv6 firewall tester installation
Before you begin please read the excellent Arch Linux Beginners' Guide.
Arch Linux Download and CF card creation
Fetch the Arch Linux download from RaspberryPi Downloads
If using win32 disk imager then please be aware that the image file needs to be extracted onto a local physical drive (e.g. C: ) rather than a network/remote drive.
Login, Change the root password and Create a plain user
Having logged in as root, then make sure you change the default password:
Then add a user: (follow the prompts)
Update your System and Install the Required Packages
Then update your system:
# pacman -Syu
Which is likely to update pacman itself – just follow the prompts and once this is complete then re-attempt the complete upgrade:
# pacman -Syu :: Synchronizing package databases... core is up to date extra is up to date community is up to date alarm is up to date aur is up to date :: Starting full system upgrade... resolving dependencies... looking for inter-conflicts... Proceed with installation? [Y/n] :: Retrieving packages from core...
Simplistic IPv6 Firewall
As a general starting point please read the ArchLinux IPtables documentation.
You will need to install the iptables modules and scripts using the following command:
# pacman -S iptables
The following IPv6 firewall is a very simplistic example, only suitable for use in a trusted environment, where SLAAC IPv6 address allocation is in place. This example is not suitable for an internet-facing Raspberry Pi where you need to include extra packet checking related to traffic sources, ICMPv6 types, arrival rates, etc.
*filter :INPUT DROP [0:0] :FORWARD DROP [0:0] :OUTPUT ACCEPT [0:0] :LOGINPUTDROP - [0:0] -A LOGINPUTDROP -m limit --limit 60/m --limit-burst 20 -j LOG --log-prefix "IPV6_INPUT_DROP" --log-ip-options --log-tcp-options --log-tcp-sequence -A LOGINPUTDROP -j DROP # Allow all loopback traffic -A INPUT -i lo -j ACCEPT # Drop all routing header traffic -A INPUT -m rt --rt-type 0 -j LOGINPUTDROP -A INPUT -m rt --rt-type 1 -j LOGINPUTDROP -A INPUT -m rt --rt-type 2 -j LOGINPUTDROP # Allow all traffic related to, or part of an established stream -A INPUT -m state --state RELATED,ESTABLISHED -j ACCEPT # Allow ping of this host to aid debug -A INPUT -p ipv6-icmp --icmpv6-type echo-request -j ACCEPT # Allow Router advertisements so we can use SLAAC address allocation -A INPUT -p ipv6-icmp --icmpv6-type router-advertisement -j ACCEPT -A INPUT -p ipv6-icmp --icmpv6-type router-solicitation -j ACCEPT # Allow neighbour adv/sol so we can talk to our neighbours (IPv6 ARP equivalent) -A INPUT -p ipv6-icmp --icmpv6-type neighbour-advertisement -j ACCEPT -A INPUT -p ipv6-icmp --icmpv6-type neighbour-solicitation -j ACCEPT # Allow SSH and HTTP traffic -A INPUT -p tcp --dport 22 -j ACCEPT -A INPUT -p tcp --dport 80 -j ACCEPT # Drop, but log, everything else -A INPUT -j LOGINPUTDROP COMMIT
Assuming your desired ruleset is stored in a file called simple_firewall6.rules then you can import the firewall rules using:
# ip6tables-restore < simple_firewall6.rules
You can view the active firewall rules, and determine the number of packets being processed by each rule using:
# ip6tables -v -n -L Chain INPUT (policy DROP 0 packets, 0 bytes) pkts bytes target prot opt in out source destination 0 0 ACCEPT all lo * ::/0 ::/0 0 0 LOGINPUTDROP all * * ::/0 ::/0 rt type:0 segsleft:0 0 0 LOGINPUTDROP all * * ::/0 ::/0 rt type:1 segsleft:0 0 0 LOGINPUTDROP all * * ::/0 ::/0 rt type:2 segsleft:0 2 208 ACCEPT all * * ::/0 ::/0 state RELATED,ESTABLISHED 1 104 ACCEPT icmpv6 * * ::/0 ::/0 ipv6-icmptype 128 116 19488 ACCEPT icmpv6 * * ::/0 ::/0 ipv6-icmptype 134 0 0 ACCEPT icmpv6 * * ::/0 ::/0 ipv6-icmptype 133 2 128 ACCEPT icmpv6 * * ::/0 ::/0 ipv6-icmptype 136 2 144 ACCEPT icmpv6 * * ::/0 ::/0 ipv6-icmptype 135 0 0 ACCEPT tcp * * ::/0 ::/0 tcp dpt:22 0 0 ACCEPT tcp * * ::/0 ::/0 tcp dpt:80 0 0 LOGINPUTDROP all * * ::/0 ::/0 Chain FORWARD (policy DROP 0 packets, 0 bytes) pkts bytes target prot opt in out source destination Chain OUTPUT (policy ACCEPT 7 packets, 592 bytes) pkts bytes target prot opt in out source destination Chain LOGINPUTDROP (4 references) pkts bytes target prot opt in out source destination 0 0 LOG all * * ::/0 ::/0 limit: avg 1/sec burst 20 LOG flags 7 level 4 prefix "IPV6_INPUT_DROP" 0 0 DROP all * * ::/0 ::/0
Once you're satisfied that the IPv6 firewall rules are performing correctly then they can be saved using the following command:
# rc.d save ip6tables
Note that if you're also using IPv4 then don't forget to setup a similar IPv4 firewall ruleset. Again this example is only suitable for use in a trusted environment and needs further additions for an internet facing machine.
*filter :INPUT DROP [0:0] :FORWARD DROP [0:0] :OUTPUT ACCEPT [0:0] # Allow all loopback traffic -A INPUT -i lo -j ACCEPT # Allow all traffic related to, or part of an established session -A INPUT -m state --state RELATED,ESTABLISHED -j ACCEPT # Allow ping of this host to aid debug -A INPUT -p icmp --icmp-type echo-request -j ACCEPT # Allow SSH and HTTP traffic -A INPUT -p tcp --dport 22 -j ACCEPT -A INPUT -p tcp --dport 80 -j ACCEPT # Drop all other TCP traffic -A INPUT -p tcp -j DROP # Allow DHCP related traffic -A INPUT -p udp --dport 67:68 -j ACCEPT # Drop everything else -A INPUT -j DROP COMMIT
This time import can be performed using:
# iptables-restore < simple_firewall.rules
Once you're happy with your IPv4 firewall then you can save the active rules using the following command:
# rc.d save iptables
Note that it is import to check full functionality still exists with your firewall in place - this particularly applies to address allocation (e.g. DHCP and SLAAC) procedures which may mean that a misconfigured firewall makes your Raspberry Pi unreachable. This is one reason why it is useful to develop the two rulesets (IPv4 and IPv6) separately, since if you misconfigure one firewall and lose connectivity then you can fall back to the working protocol version to correct your mistake. Once you're happy that both firewall sets are correct then you can insert both sets into /etc/rc.conf DAEMONS statement before the network module is called:
DAEMONS=(!hwclock syslog-ng iptables ip6tables network openntpd @netfs @crond @sshd @mysqld @httpd)
You can check for dropped/logged packets (in the examples above logging is included for the IPv6 packet filter) using the following command:
# tailf /var/log/iptables.log
Install LAMP by following the excellent Arch Linux LAMP guide.
# pacman -S apache php php-apache mysql # rc.d start mysqld
Don't forget to add a MySQL password:
# mysqladmin -u root password ‘password’ # mysql -u root -p
Then Edit /etc/rc.conf (to start MySQL at boot):
DAEMONS=(... mysqld ...)
Read and follow the Apache section - I suggest that you adjust the default DocumentRoot directory by inserting an additional directory level (e.g. htdocs) under /srv/http so that you can place other directories at this same level without them all being under the Document root:
/srv/http/ htdocs/ - DocumentRoot directory for storage of your served web pages (e.g. index.php discussed below) cgi-bin6/ - directory for the ipscan cgi executables
If you follow this suggestion then don't forget to modify the DocumentRoot setting in the apache configuration file! Having followed the php installation guide then you'll also need to update php's base directory to match Apache inside /etc/php/php.ini:
open_basedir = /srv/http/htdocs:/home/:/tmp/:/usr/share/pear/
And also define the timezone appropriately for your system:
[Date] ; Defines the default timezone used by the date functions ; http://php.net/date.timezone date.timezone = "Europe/London"
Once Apache and PHP start successfully:
# rc.d start httpd
then edit /etc/rc.conf (to start Apache at boot):
DAEMONS=(... httpd ...)
Then install the basic development tools (provides gcc, etc.)
# pacman -S base-devel
And perl and the associated MySQL interface modules:
# pacman -S perl-dbi perl-dbd-mysql
You may also wish to include spoken sound support:
# pacman -S festival festival-english alsa-utils
And update your festivalrc file:
# cat /root/.festivalrc (Parameter.set 'Audio_Command "/usr/bin/aplay -R1000 -q -c 1 -t raw -f s16 -r $SR $FILE") (Parameter.set 'Audio_Method 'Audio_Command)
And add a sound.conf file under /etc/modules-load.d to ensure the Broadcom sound module is loaded at boot. The file needs to contain a single statement shown below:
# cat /etc/modules-load.d/sound.conf
Install git, download and build ipscan
Then install git and download the ipscan source:
# pacman -S git
Then clone the ipscan source into a directory under your root user account:
[[email protected] ~]# git clone https://github.com/timsgit/ipscan ipscan Cloning into 'ipscan'... remote: Counting objects: 221, done. remote: Compressing objects: 100% (200/200), done. remote: Total 221 (delta 156), reused 85 (delta 20) Receiving objects: 100% (221/221), 102.97 KiB, done. Resolving deltas: 100% (156/156), done. [[email protected] ~]# ls ipscan COPYING ipscan.c ipscan_checks.c ipscan_db.c ipscan.h ipscan_portlist.h ipscan_web.c Makefile README sqltidy.pl [[email protected] ~]#
Now follow the instructions in the READ file within ipscan's github repository.
It's necessary to change the Makefile to reflect your Apache server's cgi-bin directory mapping:
Assuming your Apache configuration file (/etc/httpd/conf/httpd.conf) contains:
ScriptAlias /cgi-bin6/ "/srv/http/cgi-bin6/"
<Directory "/srv/http/cgi-bin6"> AllowOverride None Options +ExecCGI -Includes Order allow,deny Allow from all </Directory>
then modify your ipscan Makefile to reflect this:
# Install location for the CGI files TARGETDIR=/srv/http/cgi-bin6 # HTTP URI PATH by which external hosts will access the CGI files. # This may well be unrelated to the installation path if Apache is configured # to provide CGI access via an alias. # NB : the path should begin with a / but must NOT end with one .... URIPATH=/cgi-bin6
Make sure you have created the /srv/http/cgi-bin6 directory (or whatever you have chosen) before attempting to build ipscan. Also make sure that your MySQL database is created following the instructions in the github repository. You will need to login to mysql using the root password you previously defined (above):
# mysql -u root -p mysql> create database ipscan; Query OK, 1 row affected (0.00 sec) mysql> create user 'ipscan-user'@'localhost' identified by 'ipscan-passwd'; Query OK, 0 rows affected (0.01 sec) mysql> grant all privileges on ipscan.* to 'ipscan-user'@'localhost' identified by 'ipscan-passwd'; Query OK, 0 rows affected (0.01 sec) mysql> exit
Modify the ipscan-user and ipscan-passwd entries to use your preferences (different to the ones that you chose for root!) and enter the same credentials into the ipscan.h include file:
// MySQL database-related globals #define MYSQL_HOST "localhost" #define MYSQL_USER "ipscan-user" #define MYSQL_PASSWD "ipscan-passwd" #define MYSQL_DBNAME "ipscan" #define MYSQL_TBLNAME "results"
Then you should be able to make ipscan as root user and perform the install to transfer the necessary cgi files into your preferred cgi-bin directory:
# make && make install
Prior to running the ipscan tester it is advisable to add a cron job which will execute the sqltidy.pl script to remove the completed scan results to protect your users' security and minimise the size of your database:
First install the necessary perl mysql data base interface modules:
# pacman -S perl-dbi-mysql perl-mysql
Then modify the MySQL related entries in the sqltidy.pl script to match your chosen user, password, etc. and then ensure that the script runs standalone without any perl errors:
And finally edit the root cron job to insert the line shown below (modified to reflect your ipscan source directory):
# crontab -e
You may wish to move sqltidy.pl to another location, but ensure its permissions prevent ordinary users from reading or executing the file:
*/5 * * * * /root/ipscan/sqltidy.pl 2>&1
Check your Services and IPv6 address allocation
Now it is suggested that you check your Apache service is running using lsof:
# pacman -S lsof # lsof -i -n -P |grep http httpd 712 root 4u IPv6 1457 0t0 TCP *:80 (LISTEN) httpd 16107 http 4u IPv6 1457 0t0 TCP *:80 (LISTEN) ...
And for MySQL:
# ps -ef |grep -i mysql root 392 1 0 Aug25 ? 00:00:00 /bin/sh /usr/bin/mysqld_safe --user=mysql mysql 688 392 0 Aug25 ? 01:47:48 /usr/bin/mysqld --basedir=/usr --datadir=/var/lib/mysql --plugin-dir=/usr/lib/mysql/plugin --user=mysql --log-error=/var/lib/mysql/alarmpi.err --pid-file=/var/lib/mysql/alarmpi.pid --socket=/var/run/mysqld/mysqld.sock --port=3306
It is now worth checking that your Raspberry Pi has been correctly allocated an IPv6 address, using ifconfig:
# ifconfig eth0 eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500 inet AA.BB.C.DD netmask 255.255.255.0 broadcast AA.BB.C.255 inet6 2001:470:971f:3:ba27:ebff:fecc:dc7c prefixlen 64 scopeid 0x0<global> inet6 fe80::ba27:ebff:fecc:dc7c prefixlen 64 scopeid 0x20<link> ether b8:27:eb:cc:dc:7c txqueuelen 1000 (Ethernet) RX packets 721789 bytes 103366589 (98.5 MiB) RX errors 0 dropped 48 overruns 0 frame 0 TX packets 231210 bytes 130480722 (124.4 MiB) TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
Assuming that your Raspberry Pi has a valid IPv6 address and that your Apache and MySQL services are correctly running then you should be able to point a web browser to your cgi file:
Restrict SSH Logins
In general it is recommended that you apply all the standard SSH hardening approaches. You can also restrict logins to your newly created plain user (above) with the addition of the following line to your ssh configuration file (/etc/ssh/sshd_config). I'd also recommend that you choose an username which isn't a simple shortening of your own name:
Complete details coming soon ...
Needs to include:
- php landing page example