A NATural Progression
Netfilter and iptables make an extremely powerful firewall. But to take advantage of it, you need to master the basic syntax as explained in my first article (“Taming the Wild Netfilter”, published in the September 2001 issue of LJ), have an understanding of the modules and matches available to you and have an understanding of what a particular system can know about a packet. Armed with these three things, you can build highly complex, tailored firewall solutions for whatever problem you might have.
Know how to take advantage of new and experimental matches and targets (and always test them). You learned this in last month's article with the iptables build targets of pending-patches, most-of-pom and patch-o-matic. Testing by creating and sending specific packets to a firewall interface is beyond the scope of this particular article, but a number of utilities exist to assist you here (sendip or ipmagic come to mind).
Build the missing modules for the kernel (make sure they're selected).
Build your rule chains, more specific rules first followed by more general rules. If it helps you organize things, go ahead and build custom user chains that can be called from another chain. While this was not covered specifically in this article, it was addressed in the September 2001 article. Use everything at your disposal, including the LOG target to help you see if particular rules were applied and to which packets.
With just some basic knowledge, iptables are not difficult to use. Read some iptables scripts on the Internet. I don't recommend using them as they are; they almost certainly won't work for you without a lot of tweaking, but they will show you syntax, rules (from which you can grab fragments), thought processes, etc.
- Readers' Choice Awards 2013
- Mars Needs Women
- Linux Kernel News - November 2013
- Sublime Text: One Editor to Rule Them All?
- RSS Feeds
- Raspberry Pi: the Perfect Home Server
- December 2013 Issue of Linux Journal: Readers' Choice
- Tech Tip: Really Simple HTTP Server with Python
- IBM Will Minimize Impact of Future Disasters
- Linux Systems Administrator
- The kernel doesn't really
1 hour 47 min ago
2 hours 18 min ago
2 hours 18 min ago
4 hours 23 min ago
- This should be very helpful
5 hours 37 min ago
- As much as I share your point
7 hours 57 min ago
- So girls had it better ?
11 hours 28 min ago
- Reply to comment | Linux Journal
11 hours 48 min ago
- why is GNOME 3 in the fifth position at 14.1 %?
17 hours 21 min ago
- Sublime Is Brilliant!
22 hours 24 min ago