By Mike Mullins
Every network device on your network has some type of logging capability. Switches and routers are extremely proficient in logging network events. Your organization's security policy should specify some level of logging for all network devices. Your access lists surely contain the log command for all denied ports and protocols.
It's important to deny traffic you don't want in your networks, but you also need to know who's sending that traffic. Some resourceful hacker could be hammering away at your outside interface and eating up bandwidth and processes. You need to know where that traffic is coming from.
But the truth is that admins typically don't log routers and switches. When a problem occurs, we just reboot them or bounce and interface, and then chalk it up to a hardware glitch. Don't go another day without setting up a centralized logging server.
Routers and switches will send log traffic on UDP 514 in a syslog format. It's just a matter of providing a secure platform to collect that information. I recommend setting up a Linux box to handle the traffic. It's simple and inexpensive, and it provides data security to some of the most valuable information about your network.
When implementing a central logging server, you need to take a number of key steps:
These six simple steps will go a long way toward providing a secure and central location for all of your networked devices' log files.
Although setting up a logging server gives you another machine to manage, you gain a central point from which you can now read all of these valuable files. Before you begin pointing routers and switches to your new security device, prepare your logging server first.
Once your logging server is up and running, you'll have no excuse not to read your logs.
Mike Mullins has served as a database administrator and assistant network administrator for the U.S. Secret Service. He is a Network Security Administrator for the Defense Information Systems Agency.