It's also important that one avoid:
* The faulty assumption there is but one problem
Here's an interesting example that I came across several years ago. It was in an office with lots of PCs plugged into RJ45 10baseT ports near each desk. One PC had lost connectivity. I came and checked that the software was installed and running. Probably did something like ping 127.0.0.1 to satisfy myself that it wasn't a problem on the PC itself. Then I unplugged the cable from the RJ45 port in the wall and tried another port. It still did not work. I swapped in a new cable and it worked fine. Most people would stop right there, but I followed up and tested the existing cable in the lab. It worked just fine. Why did it not work before? There must be some problem with the switch or the wall wiring and somehow two RJ45 ports did not work. After a bit of poking and discussions with the employee at that desk, it turned out that the cable lay in a bad spot and often got caught on her foot as she rushed off somewhere. It turns out that the little metal pins inside the RJ45 socket had been bent. It was just sheer luck that swapping the cable caused contact to be made again. And the second socket was also bent. When that one ceased to work the employee had swapped cables themselves. The real solution was to replace both sockets and install a longer patch cable that could be placed where feet would not get caught up in it. Troubleshooting is made easier by methodically doing the work and following through. If I had not had the lab handy I probably would have swapped the "bad " cable back in to verify that "trouble" accompanied the cable. But it is also easier to troubleshoot when you have a stock of interesting war stories in your memory to encourage you to "think outside the box". It's the blend of creativity and methodical work practices that makes a good troubleshooter, technical or otherwise. --Michael Dillon