Firewalls monitor and regulate the data moving on and off your computer or network. They can keep criminals out while allowing legitimate network traffic in. Mac OS X comes with not one but two firewalls of its own. However, those two aren’t always enough.

The Threat

Years ago, a bug (long-since fixed) let attackers send Macs a so-called “ping of death”—specially designed network traffic that could crash a system. There aren’t any such network vulnerabilities on Macs (that we know of) now, but many of Apple’s security updates specifically address network vulnerabilities. Clearly, Macs aren’t inherently immune.

With millions of computers in the world, it might seem that the odds of your Mac being targeted are awfully small. But there are computers out there that do nothing all day but probe Net-connected machines for vulnerabilities; it’s certainly possible that one will find yours. And don’t forget that any time you’re on a network—a coffee shop’s Wi-Fi system, for example—you’re exposed to anyone else on that network.

The risks—the loss of private data and the hijacking of your Mac’s computing power—are great enough, and the cost of prevention low enough, that implementing a good firewall on your Mac and your local network is a no-brainer.

OS X’s Firewalls

All versions of OS X through 10.4 (Tiger) have included a Unix-based firewall called ipfw. In security parlance, ipfw is a packet-filtering firewall: it checks each packet coming or going through the Mac’s network interfaces against a set of rules, and allows it to pass or blocks it.

Packet-filtering firewalls like ipfw classify network traffic two ways: by type, using port numbers, and by origin and destination, using IP addresses. For instance, a packet-filtering firewall could accept file-sharing connections from IP addresses of your work network but not from other addresses on the Internet.

Source: www.macworld.com