The /usr/bin/eject utility on a Linux system is a good way of figuring out which machine has what name in the KVM when you're dealing with a rack of machines, many of which are likely to be mislabelled. Use the eject utility to open the CD-ROM drive and see what machine you're really connected to; then update the labels in the KVM's configuration appropriately.
If your fibre-optic network card isn't seeing any traffic at all, try switching the plugs on the card. Some optical network cables don't have colour-coded connectors so it's easy to plug them into the wrong jacks on the card.
Supervisors and ninjas have a lot in common. Both are capable of god-like feats of stealth, obsevation, and scaring the living hell out of you if and when you actually do see them. Be nice to both.
No matter what happens, end-users will always blame the firewall when they can't reach a server. Even if the server's sitting next to them and on fire, they'll still blame the firewall.
Redhat Linux versions 8 and 9 have a nasty bug in them - sometimes if you're trying to remove or install an .rpm file the RPM utility itself will lock up, which means that the database files RPM uses to keep track of what packages are installed and what version they are are stuck in a locked state. The solution to this problem is to delete the locked temporary database files and reconstruct the database (rm -f /var/lib/rpm/__db* ; rpm -vv --rebuilddb). Yes, this looks nasty - you're deleting system files. But it's nothing that the RPM utility can't rebuild if you give it a few minutes.
It should be noted that you should use the top command to see what processes are using the most CPU time. If rpm is at 100% then it's actually doing something and you shouldn't kill it. Yes, wait a few minutes, then kill it and then reconstruct everything.
Replacing corrupted packages in a Redhat Linux box is as easy as rpm -e --nodeps packagename ; rpm -Uvvh /path/to/replacement.rpm.
To access the BIOS configuration on an Acer Aspire, use the key combination control+alternate+escape.
Trying to get Postnuke, the PHP-based content management system running can be daunting because there's next to no documentation. You can find an early draft of the user's guide here - print it out and save it.
In case you're having trouble figuring out how blocks work, here's how you do it: Follow the instructions in the archive to install your new block package. Go to the Administration page, select Blocks, and then click on the New Block link at the top. Type in the title of your new block, then click on the Block field and pick what kind of block to make it. Select the new kind of block that you just installed. Click the Commit Changes button. This will take you to the new block's configuration screen. Set it up however you like. Click the Commit Changes button. On the next screen use the up and down arrows to push your block around on the sides of the website. There you go.
It's not production until you send out the memo asking everyone to use it. Go ahead and experiment with it, no matter how lame you think it is. Just be sure to have a backup someplace.
Any file manipulation that will take you several dozen lines of shell script you can probably just as easily do in four or five lines with awk.
You can cut down many false alerts when using Snort as your intrusion detection system if you define the EXTERNAL_NET variable in snort.conf as !$HOME_NET, meaning that anything not in your home net is outside, and therefore to be watched closely.