pibara / cduck Goto Github PK

Capibara Distributed Cloaking Kit (cduck) 0.9.6 beta

This package contains the software needed to turn your static-ip dsl/cable
*nix system into a node of a distributed URL cloaking server.
The functional software exists of 4 components:

CapDNS:   The Capibara DNS server, this server will answer "that's me" 
          to most relevant address queries.
CapDB:    The database part of the url cloaking system.
CapHTTP:  The Capibara HTTP front-end to the database server. 
CapCron:  The Database and DNS synchronizing cron script.
SnowBall: A wrapper script that exists only for the reason that
          the CapDNS should not give "that's me" answers if either
          of the other processes is down. Snowball will make sure
          all the cduck processes die if one of them dies in order
          to ensure the integrity of the distributed system.

Next to this cduck takes a set of  specific measures for security. 

Some parts of the containment are platform specific and/or very 
experimental and likely not portable, for this reason the containment
part of the installation is optional. 
You should read CONTAINMENT for more information on the security 
measures build into and around cduck, and on how to use them to the 
fullest on the Linux >= 2.4 platform, the FreeBSD platform and the
Solaris platform. If you are successful in running 'make containment'
on any system other than the three systems named here, or need to take 
undocumented actions in order to get it up and running on any of the 
named systems,please contact the author with specifics.

To install the server please first take the following actions:

CHECK PREINSTALL CONDITIONS:

In order to completely run all components of cduck, you will need one of the following:

* A Linux system with 2.4 kernel and iptables 	(http://www.kernel.org/)
* FreeBSD system with ipfw and fire-walling enabled. (http://www.freebsd.org/)
* Solaris system (http://www.sun.com/)


Further you will need:
* perl 5.6 or better 		(http://www.cpan.org/src/index.html)
* libwww-perl 			(http://www.cpan.org/modules/by-module/LWP/)
* libpcap			(http://www.tcpdump.org/)
* The Net::Pcap perl module	(http://www.cpan.org/modules/by-module/Net/)
* Berkeley DB			(http://www.sleepycat.com/)
* The DB_File perl module	(http://www.cpan.org/modules/by-module/DB_File/)

You might be able to run cduck on an other *nix operating system. 
This has at this point not been tested to any extend. 

Please make sure you have all of the above installed and working on your system 
before proceeding to the installation of the containment measures. 
The creation of a proper chrooted environment for cduck relies on this, and if it is 
not installed at the time that 'make containment' is run it will result in a non working 
installation of capcron.

UPGRADING FROM PREVIOUS VERSION:

If you already have a running version of cduck on your system, than you can
make use of the following procedure:

* run 'make save' to make a backup of your basic setup files.
* run 'make uninstall' to uninstall the complete capibara tree.
* run 'make install' to install the updated capibara tree.
* run 'make rest' to place the backup-ed file into the new tree.
* run 'make fetch' to get a recent version of the map and zone files.
* run 'make containment' (optional) if you wish cduck to run chrooted.

CREATING THE USERS AND GROUPS:

You will be able to use 'make users' on most Linux systems, it installs users 
and groups. I don't quite know how portable it is, and you may wish or have to do the
steps manually (it is suggested to do so at least on any non Linux system for now) 
if you don't like install scripts messing with your /etc/ stuff, or if the script 
tells you it's to stupid to work on your system.

INSTALLING THE SOFTWARE

* run 'make install', this should copy the files to the right
  place, and set the correct ownerships and acl.
* run 'make conf' in order to do some node specific configuration.
* Place the appropriate zone and map files on a reliable secure web server
  and edit the /usr/local/capibara/etc/cduck.conf file. 
  See 'man cduck.conf' for more info.
* Run 'make fetch' to fetch and create the database and zone
  file for the first time. 

INSTALLING CONTAINMENT

You are suggested to read the file CONTAINMENT for the configuration
of the advanced security measures for cduck.
If you feel confident you might type 'make containment' and take close
note of any messages it returns.

RUNNING THE SOFTWARE


* Test the configuration by running 'rc.d/cduck start' and (on systems with 
  containment installed) 'rc.d/tids start' to see if
  everything works OK. Is this works OK, than move these scripts to the
  appropriate places in your rc tree in order to have the services
  started up at system boot. 

* run 'crontab -e' (as root) and add the following line :

'11 3,15 * * * /usr/local/capibara/bin/capcron.pl'


If you have problems running the software, please first consult the file FAQ.draft.

I've only tested it on a limited amount of systems, so there might be some 
things that don't work out of the box yet for everyone. If this is the case 
please report any problems or suggestions on http://sourceforge.net/projects/cduck/
or e-mail me with details.

Rob J Meijer <[email protected]>

Until the current release (0.9.6) the cduck user-base seems to be rather fragmented. Most use
seems to consist of single nodes, and I know of only a few small closed groups of users using 
cduck within a particular geographic region.
There would however IMHO be large advantages to using cduck to establish redundancy across
network-technical or geographic regions. If you are interested in becoming part of a 
cduck cluster please consider posting on the sourceforge forum:

http://sourceforge.net/forum/forum.php?forum_id=135728