Well, being the dedicated type that I am, I wrote a script over my Thanksgiving break at my Mom's house. It worked great. It would scan through the hosts file for interesting hosts, query the hosts for active information, and build an MRTG configuration around it. Some of that can be done with existing indexmaker, but I had coded in configurable variables so that it could accommodate ports that were up/down, up/up, or down/down for various amounts of time and keep the data or shred it. Know what I did with it earlier today? I deleted it. Completely accidentally... deleted. I felt ill. There's no "recycle bin" in Unix. There's no "undelete" command.

Read more »

One weapon every network ninja should have in their belt is ACL knowledge. This tool can be used for everything from tweaking routes to selecting traffic for blocking, modifying, or accepting. Let's run through some of the methods and list some examples.

Read more »

ICMP, the Internet's best friend. Testing the paths, passing back blocked path info, or warning of packet size issues, this power-protocol get's two powerful Ninja thumbs up. Ping and traceroute are the so embedded into basic troubleshooting that often-times non-technical staff members at least know to send them along with any requests for troubleshooting. Ping and trace can be used to acquire more in-depth insight than basic reachability info.

Read more »

This may not be pervasive in your environment, but I've asked for this email from people and received it in the form of screen captures. I kid you not, ragged-looking, email-size-inflating screen captures that you stand no chance of using the "copy" functionality on. It's not really their fault; they likely asked someone how to get the information out of a command prompt once and that was the answer they were given. Take the time to educate them as to the proper way to harvest the information with the following steps.

Read more »

Whether you're a developer, firewall engineer, server owner, or a network engineer, having precise knowledge of what is passing in and out of an interface can be a lifesaver. Don't get me wrong, Netflow is awesome and graphs are too, but whether it's break/fix, new functionality, proving it's not your device that's broken, or even that someone else's is, when you need to definitively prove/disprove something, packet-level perception is where it's at. Whether it's accusations of latency, malformed streams, packet drops, or NAT issues, if you're looking for liberation, you've found it.

Read more »