This morning I was doing a fresh install of Windows XP SP2 into a Virtual Machine. So far, things are fine I went through windowsupdate and found 3 updates the first time, then rebooted and hit windowsupdate again to see 55 updates available. A lot of times when I set up a new pc for somebody they wonder why I want to check windows update multiple times. They’ll usually say something along the lines “but it’s brand new there shouldn’t be any updates.” Well, this install was from a SP2 disk and there have been a large number of updates since that was released. Many manufacturers use fairly sophisticated techniques to roll out the default install images they use, but it’s still very possible that your machine will have several updates waiting for it when you get it.

When I go through and update a machine, my usual practice is to check windowsupdate until there are no “critical updates” left. So, with this mornings VM install, I checked once, 3 updates, now I’ve checked and seen 55 updates. I’m not going to assume that is all. After it reboots I’m going to REvisit again and see if anything is left. “Well can’t you do automatic updates?” yes and they’re enabled. I just 1) don’t want to wait for them and 2) want to make sure that I’ve got everything up to current as soon as possible. After things are “caught up” then automatic updates should do ok keeping up with new releases from today forward.

Don’t assume because it’s brand new that it’s current on all it’s updates. This applies to more than just Windows, 3rd party software included (Mac and Linux systems as well.) Even hardware. Now, I’m a bit less likely to encourage keeping up to date with the firmware for hardware devices like routers and wireless access points, but… Many times those firmware updates will resolve problems you’ve seen with the equipment and there might be the occasion that a security update is needed for those pieces of equipment too.

This is one that’s not as obvious as it should be. Every keyboard has a “print screen” button – usually it’s in the group over above the insert/home/pageup/delete/end/pagedown block. Sometimes (especially on laptops) it’s shared with something else. But…. most people will notice if they press “print screen” on a windows pc… not much seems to happen. (On my KDE/Linux desktop it pops up a program called ksnapshot…) If the print screen key is shared with another function (say on a laptop) you may need to press the function button at the same time as the print screen button to get the following to work.

One way, you can check if you do need to press the function (fn) button at the same time is to look at the color coded text on the fn button and where the “print screen” text is. IF they’re the same color, press the function key and hold down while you press the print screen button. Of course, if you’re not using a laptop… ignore most all the above and press the “print screen” button. uhmmm nothing happened. Ah, but it did… windows just took a “snapshot” of everything you see on the screen and is saving it for you. All we have to do now is find a place to put it.

Click on the start button and go to “all programs, accessories, paint” (or you can also go to run and type mspaint and click ok). Click the edit menu and then click paste. You should see your “screenshot” appear in the paint window. The screenshot that windows does is not very fancy, it just grabs the entire screen when you press the print screen button. (Then you can paste it into whatever image editing program you use.) After it’s been pasted into mspaint, you can print by going to file, print. (Of course, other programs may give you fancier options for dealing with your brand new screenshot.).

The Ksnapshot program I use under linux is a bit fancier in that it can capture just the active window (firefox maybe), or the entire desktop, or it can take a screenshot after a user-defined delay. I’m sure there are addon programs to windows that may make this process easier. But now you know how you can make a screenshot under windows xp to print what’s on the screen using the print screen button.

How many times have I had this kind of phone call…. “my computers not working”… “the light on the monitor is on, but there aren’t any lights on the keyboard. everything seems connected, but there’s nothing on the screen.” Usually, my first question is along the lines of… “have you tried turning it off and back on.” I sometimes hear “I can’t use the start button” or something like that. At this point, I suggest that they try disconnecting the power and letting it sit for a few moments (5-10 seconds is usually enough.) Plugging it back in and trying the switch.

It’s kind of amazing that I still get calls like that, but it is basic troubleshooting. Unplug all power from it, let the capacitors drain (that’s why we let it set ~5-10 seconds) and try plugging in and flipping the switch again. Check the outlet, has a fuse blown? Do the lights on the PC come on?

In the most recent case, flipping the switch (ok PUSHING the switch) was all it took to get things booting again. I don’t know how it went out, but if things come up normally, it probably isn’t a big issue.

While I’m on the topic of simple fixes…. I can’t begin to count how many times “our server is down” has turned into “maybe the network cable got bumped when you moved your desk?” Check the connections, reboot, try again. In some cases, the best way to check a connection is to unplug/replug…

This is going in the basics category. Quite often I visit a computer that has a mouse which is just a trial of patience to use. I mean, you have to pick it up and move it, shake it a bit, turn it over and clean out the lint from the “ball” area…. The good news is that mousing does NOT have to be like this. If your computer has a PS2 (small round plug) for your mouse, or a USB (small rectangular plug) for your mouse, it should be VERY easy to replace that old ball style mouse with an optical mouse.

Some people get optical confused with wireless….. an Optical mouse basically uses a little light (instead of a roller ball) to track the mouse’s position. The advantages are: you can mouse on most any surface (that’s not reflective…); you don’t have to clean dust out of the tracking; they offer much smoother “mousing”.

I personally like using an optical mouse with my laptop. USB makes it fairly easy to hotplug and use it… (PS2 mice have to be plugged in when the system boots to work.) The days, I will not buy a trackball mouse for my own use, or for a customers new system unless they absolutely refuse the optical. (Older systems that use a serial port \_/ shaped connector are unfortunately out of luck, I don’t know of ANY optical mouse that fits the serial port. Fortunately, that is much older hardware – around vintage Windows 95-ish…)

Optical mice can be wired like a traditional mouse, or wireless.

Wireless mice are good options as well, there are a couple things to think about. Many wireless mice have a transmitter that plugs in through USB. Some are about the size and have the about the same length cable as a mouse itself, so if the goal is reducing clutter, shop around for a wireless mouse with transmitter that just plugs into usb and doesn’t straggle far… The big thing though is to keep batteries stocked, because the wireless mouse HAS to have power, it will typically use AA batteries. It’s a good idea to keep spares on hand. If you want, you should be able to use rechargable batteries although I haven’t tested that out. (Rechargable batteries put out a bit less voltage than traditional alkalines…)

Of course, with a wireless mouse, if it suddenly stops responding, the first thing to check might be the batteries. Also, many times there will be a button to “find” or “page through” the radio frequencies that it can use to communicate with the transmitter, so that’s worth trying as well.

A lot of places sell mice for laptops. Many times they’re smaller and there is a bit of a premium in price for those. A “desktop mouse” will work just fine if you’re comfortable with the size. In fact, the one I carry for my laptop is an identical Microsoft Wheel Mouse Optical to the one I have on my desktop. I’ve never got used to the smaller size of the “laptop” mice….

Anyway, those are just a few ways you can upgrade your mouse and give you a less frustrating mousing experience.

Dealing with text…. we’ve looked at a couple of basics on logging in and starting to use the command line and hopefully not feeling too helpless there… we’ve seen ways to navigate directories and how to find new commands and how to find out more about how to use them. Now it’s time to talk about how to manipulate (work with) text files. Most every important configuration file in linux is text based and editable from the command line. This can be a VERY good thing at times, or very intimidating if you’re not comfortable with a command line environment. It’s worth noting that you CAN edit configuration files with a graphical interface text editor….

So, there are a number of text editors, one that I use quite a bit is pico, but there is also vi and emacs. pico is a nice, easy to use text editor and is highly recommended for beginners. With pico, ctrl-o writes a file and ctrl-x closes, with vi it’s a bit more complicated, esc must be pressed to quit editing and then :wq are typed (and enter) to write, and quit. vi is worth an article or two on it’s own and frankly, I don’t know that it would qualify for “basics”…. Let’s assume we can use pico…

First off, let’s take a look at a text file…
$cat textfile.txt

should give you a stream across the screen of the contents of textfile.txt… nice, but hard to read. OK, try this…
$less textfile.txt
now you can use the up or down arrows to move up or down one line at a time, and page up/page down to go down a page at a time.

$more textfile.txt has the same effect… except “enter” moves you by line and space moves by page…

What if you wanted to find out if there was a certain word in textfile.txt…..

$grep specialword textfile.txt
this line uses specialword which is a keyword that we will later search for

grep allows you to search for a term…. I use it daily….
For instance… (Has the googlebot visited?)

grep googlebot access_log -i
-i tells it to not be case sensitive…

It’s possible too, to chain grep commands: what if I wanted to see if the googlebot had visited pages published in January…. (2006/01 in the path…) Us the | (pipe) to feed output of the first part, to whatever followes the | (the | is usually the “shift” equivalent of \ )

grep googlebot access_log -i | grep 2006/01

I hope you’ve enjoyed the last few articles which give some ideas for how you might get started with the command line if you’ve never tried it before… Fortunately most operating systems respond to the “help” command and many have features to give further information when you follow a command with -h or –help or /? or /h …….

Dealing with text files at the command line can quickly get very difficult and take us out of “basics” territory into lower-intermediate levels, so I’m going to cut this series here before I get you in too deep…

Now that we’ve logged in and learned how to move from one directory to another (and find where we are if we get lost…) it’s probably time to start learning how to find out about other commands. Well, if you press tab at the command line you will likely see something that starts like this….
$
Display all 4601 possibilities? (y or n)
and if you answer with a “y” you will see a long list of programs that could be run. That’s not usually the best way to find a program though.

One of the things I like in linux is good console based help features…. let’s say I want to burn a cd, but I don’t know what tool to use…. $apropos cd

I get a long list… but one entry that might catch the eye is ….
cdrecord (1) – record audio or data Compact Disks or Digital Versatile Disks from a master

okay, so how do we use it?

“cdrecord –help”

will tell us a lot, but might not be as helpful as….

“man cdrecord” which will give us a “manual” page as well….

most commands like cdrecord watch to see what “options” have been requested of the program. In the above example –help is an “option” –help is usually used to list what options a program accepts and gives a brief explanation of what each of those does.

If –help doesn’t work, you might try -h (windows tends to use /? ) Many times if you just type the command name without any options (and it expects something) you will get a brief listing of the most commonly used options (options are sometimes called “switches”)

This is part two in a series of “how to use the command line in linux” style articles…. these are intended to be quite basic for those that have not used a command line before…. In part one we logged in (if necessary) and found out about the help command, navigated a bit with the arrows and exit -ed from our login. This time we’ll try to learn a bit more about dealing with files and navigating. In a graphical interface, if you browse files, you’re presented with a list of files and folders for a directory that you can scroll through and read at your leisure. In the command line we have to tell the computer to list the files and folders for the directory that we’re currently “in” to list those in linux, we use the ls command….

When you log in under linux and open a command shell, you are usually put in your “home directory” (/home/username/)

$ls
10-23-05-username.xpm dead.letter Documents/ potatohead122804.tuberling music.raw tmp/
9-20-05-.xpm Desktop/ Download/ Music/ Pictures/ Video/

So…. let’s see, items with / after them are directories, everything else we see is a file.

how should we move to a folder to look? we need to “change directories” don’t we? cd is the command to use…

cd Documents (remember linux is case sensitive…) and press enter…
now ls
document.txt

there are some handy shortcuts too… to move “up” a directory type cd .. and we move “up” in the hierarchy (through nested folders) by one folder. We could also go directly to a folder by specifying it “explicitly” “cd /home/avery/Documents/web” for instance would take me to that folder no matter where I currently have the console open to.

If you get lost, you might type “pwd” to see the “present working directory”

I know many people get intimidated with a black screen, white text and blinking cursor. No mouse, not knowing where to start…. well, I’m going to spend a bit of time in this article trying to help show you how to survive the command line in linux. For starters, don’t think that linux is a command line only operating system. In some installs, the administrator may not want a gui, but there are some VERY functional and full featured graphical interfaces for linux. For me though, if you know what you want to accomplish, the command line can be a much more efficient way to “get the job done” for many tasks.

Ok, so let’s start…. we’ve either opened up a console window (konsole if you use KDE), or we have a fullscreen console open. IF we’re staring at a full screen console you may need to log in… type your username and press return (no mouse for clicking ok here…) and then type your password (likely it won’t “echo” or print on the screen as you type.) Press ok when you’ve typed that in. So now you have a very spartan looking $ prompt and blinking cursor… now what?

For starters, if you’ve never been at a command prompt before… you might type help and press enter. You’ll likely see a list of commands – when I do that… I see a lot of commands but near the top it says….

Type `help name’ to find out more about the function `name’.
Use `info bash’ to find out more about the shell in general.
Use `man -k’ or `info’ to find out more about commands not in this list.

Ok… so we can see some basic commands and have an idea of how to get a bit more information (help history for instance would give help on the history command.)

There’s also a hint in the above at other ways that we might find help (on a great many programs) The man and info commands. If the text has scrolled down too quickly to read, you might use shift-pageup to “scroll” back and see what you missed. Of course, if you’re using a console window you should be able to use the mouse to scroll.

For the most part when you have a console, the mouse is not used. If you need to move the cursor left or right, try the arrow keys and of course, as we saw for login, the enter key is like pressing the ok button with a mouse….

To log out, simply type exit (and press enter)

I did an article a while back about cron and scheduling tasks in linux (I also mentioned kron which is a graphical interface for cron scheduling…). That works well and good for things that happen on a recurring basis, but the next question is what if I want to schedule a command (or series of commands) to happen once and not again? Well, to tell the truth I’ve used cron for this on occasion, being sure to revisit the crontab and remove the entry before it repeats again, but there is a better way. At.

I could probably do a bad vaudeville-esque bit and say “at what..”, but there is a very powerful and simple scheduler called at in linux. If you open up a command shell and want to run the command “killall mplayer” to end all running mplayer processes at 10 pm, here’s how you’d do it…..

$at 10 pm
warning: commands will be executed using (in order) a) $SHELL b) login shell c) /bin/sh
at> killall mplayer
at>
job 12 at 2006-01-16 22:00

So, I typed “at 10 pm” (without the quotes) a the $ prompt… it gave me an interactive way of entering commands…. next I typed “killall mplayer” (and pressed enter), then since I didn’t have another command I wanted to schedule, I merely pressed ctrl-d to exit and was promptly told that the task was scheduled for 2200 hours tonight (which is 10 pm).

at can handle fairly human-friendly time scheduling…. at 5 tuesday for instance would schedule for 5 am on tuesday (today is Monday and so it schedules for tomorrow, but it is swift enough to figure that if the time has already passed for the week, then it will do it the next time it comes around.)

atq can show what’s scheduled…
$ atq
13 2006-01-17 05:00 a avery
12 2006-01-16 22:00 a avery
$

atrm can remove scheduled tasks…. atrm 13
atrm 12
for instance removes the above entries….

it’s also possible to feed at a file with a list of commands…. at -f listofcommands.txt
the format is usually one command per line. More information on at is available by typing “man at”
(again without the quotes) at the command line.

This is one that I run into from time to time. Most laptops the last year or two (actually maybe a bit longer), have built in wireless networking. This is great and makes for easy access to wireless access points just about anywhere. In fact, most current operating systems (Windows XP included) usually do a great job of detecting available networks. (Sometimes they do too good a job which is why wireless security is an issue, but we’ll save that for a future post.) I’ve run into several laptops over the last few years though that supposedly had wireless networking, but the user hadn’t been able to access any wireless networks with the laptop. Usually it’s an easy fix….

That’s why we’re in the “basics” section right? It is an easy fix, because many laptops that come with built in wireless ALSO come with a switch or button of some sort that turns it on and off. Thankfully, the default setting for most of these I’ve seen is off. That’s good, because you wouldn’t want your laptop networking with other wireless devices without being aware of it. So, if you’re having problems getting wireless networking to work on a new laptop, look around the edges of the case and see if there isn’t a small switch or button that might make a difference.

When it’s switched on it may take 30 seconds or so for Windows (or whatever your computer runs) to detect the change and show available networks. If encryption is enabled, this is where we’re looking at another story, but for “in the clear” networks, you’re probably already online.

What’s really disturbing is that one customer asked at the place they bought the notebook. They asked one of the sales people at Best Buy how to use the wireless networking on the notebook. They didn’t know and suggested that maybe the customer had to use a verizon wireless (cell-phone modem) card.