When we think of data protection, most people just imagine themselves copying their computer over to another device on a regular basis… so that it can be copied back in an emergency. Of course, it’s not that simple.

There are many different challenges that make data protection difficult, and many different recovery scenarios need to be accounted-for:

• How would you react if you accidentally backed up a virus?
• What do you do if an employee accidentally deletes a file or saves over an important document?
• How fast can you get back up if your system goes down during business hours?
• What do you do if you have more data than you can reasonably back up?

These are just a few of the many data protection scenarios that you need to prepare for. In order properly defend yourself, you need a solution that is simple, modular and capable of handling a wide range of challenges.

The methodology I prefer involves 3 different backup processes which complement each other very well.

When you first set up your computer, you’ll want to install all of the applications you need and create a recovery image for this machine. This will come in handy if you ever need to perform a bare metal restore.

Restoring your system from a disk image, with all of your favourite applications pre-installed, can save you many hours when re-building your machine from scratch.

You’ll want to keep 2 copies of this system image. One copy should be kept on-site, and the other should be taken to a remote location for safety.

Of course, when it comes time to recover, this disk image will already be several months old. Also, it will only contain applications…. not data.

What you truly care about the most is your irreplaceable data. All of those pictures, Word documents, and Photoshop files are very important and valuable to you. At the very least, you’ll want to back these up on a daily basis.

Also, you’ll want to maintain several historically versioned copies of all your data backups. This way, you can still recover your files if you accidentally back up a corrupted version. I recommend keeping about 30 days of historical backups… just to be safe.

And just like your system image, you should keep 2 copies of each backup… one off-site, and one close-at-hand.

As you keep using your machine, you will accumulate a lot of junk that can quickly fill up your hard drive. Although you may not need any of this data right now, you might require access at a later date. This type of clutter can cause your backups to take much longer, and greatly extend recovery times.

In order to optimize your backup process and improve the performance of your machine, you’ll want to periodically clean up your hard drive and send inactive files off to an archival storage device. For this kind of long-term storage, I’d recommend a stable & crash-resistant media such as online storage or backup tape. (External hard drives have too many unstable moving parts)

A good policy is to archive any files that have not been accessed within the past 6 months. This should be enough to clear 60-80% of the space on your hard drive, and also improve backup & recovery times by 60-80%.

If you ever need these files again later on, retrieving them from the archives should be a simple matter.

If your computer were ever to suddenly crash, you would first load up your system image to restore your machine and applications to their previous state, then you would load up your most recent backups. If you need any additional data that wasn’t included in the backups, you
can selectively grab only the files you need from the archives without having to load the rest.

Now you see how much easier backup and recovery can be when you’re organized. And with this approach, you’ll also be prepared for a number of other unexpected data loss scenarios.

About The Author: Storagepipe offers online backup services that are designed to simplify backups, and speed up recovery.

There’s only one thing worse than having no backup at all: thinking you have a backup, and finding out it doesn’t work when you go to restore.

It seems like common sense, but unfortunately backup validation is often overlooked. After establishing a backup system, it’s vital that you periodically check to make sure that it is working and that you are able to restore in the case of a disaster. A system administrator that has configured a nightly backup but has never performed a test restore hasn’t done his job; he can’t say with any degree of certainty that he’ll be able to recover lost data in the event of a disaster, because he’s never tried.

There are many reasons why a backup might turn out to be a dud, even if you thought it was functional. Here’s a quick list, including things that I’ve either seen personally, or that I’ve heard about second-hand:

1. Your backup restores, but you don’t have a few vital files or directories it turns out you really need.
2. The backup used to work, but at some point the media became full, and backups haven’t been working for some time.
3. The backup used to work, but at some point the media became corrupted, and you can’t restore.
4. The backup target is a remote network host, and the SSH key changed. The backup script has been failing to login since then, and you don’t have any recent backups.
5. Your backups are working, but you aren’t backing up often enough to save everything you need in the event of a disaster.
6. The backup target is an NFS share, which at some point became unmounted, and you’ve just been making second copies of everything on the local hard disk — not very useful.

Notice that in all of these cases, the very first backup was completely successful, but over time something fell apart. Therefore, it’s not enough just to check your backup once; it’s vital that you periodically check to make sure everything is working and complete. While many of these cases can be handled by proper backup software (corrupt media, unmounted NFS share, etc…), others are things that require human intervention (such as making sure you’ve got all the files you need, and that you back them up on a proper schedule).

Additionally, when you periodically perform a restore test, you gain:

1. Confidence that the backups are good, and that they are dependable in the event that you should need them.

Peace of mind is priceless.

2. A tested backup procedure.

Having a data-loss event on your hands and having to restore without a procedure to follow is nerve-racking. If you’re in charge of restoring data for a company, you should have a written procedure to follow. This way, not only will you know exactly what needs to be done, but it will be possible for others to restore in the event of an emergency. The procedure should be written in such a way that someone familiar with the backup system can follow it to restore the vital data if necessary.

So if you haven’t already, go now and make sure that your backups can be restored, and that you have everything you need.

What’s your backup system? Do you periodically make sure that your backups are functional, and you’re backing up everything you need to have in the event of a disaster? Are you sure that your are safe in the event of system loss? Share your backup and restore method in the comments below.

Absolutely brilliant photo credit goes to Matalyn.

1. Installing xplanet

The first thing you’ll need to do is install xplanet. This can be done quickly and easily from the command line:

Ubuntu: sudo apt-get install xplanet
Fedora: sudo yum install xplanet

2. Running from the command line

Xplanet has lots of options, which you can see in its man page if you run ‘man xplanet’. I did find that the manual isn’t entirely up to date, however – it doesn’t list all the map types that xplanet actually supports with the -projections flag. Nevertheless, the manual is still a good place to start if you’re looking to explore xplanet’s featureset.

You can run xplanet from the command line to get an idea of what kind of map you want to generate, how large it should be, and what options you want to use. For example, you’ll probably want to generate an image that fits your screen resolution. Use the -geometry option to specify the size you want, in pixels. You can use the -window option while you experiment, until you get the map exactly how you want it:

xplanet -projection rectangular -geometry 1360×768 -window

The above are the options that I use, but of course you should play with it until you find something you like. One thing you’ll probably want to tweak is the -projection option, which determines what kind of map you want to generate. Also check out the -latitude and -longitude options to tell xplanet where to center the map; you can set it so your location is at the center of the image, if you want. Check out the man page for all the details (there are many!). To help get you started, here are a couple of fun projections and the commands to generate them:

A zoomed-out globe with New York at its center:



xplanet -radius 30 -num_times 1 -latitude 40 -longitude -73 -projection orthographic

You can view more than just the Earth; here’s a spectacular view of Saturn!


xplanet -body saturn -latitude 25
3. Setting the background

Once you’ve picked out the options you want to use for your background, you need to tell xplanet to save it to a file so you can select it as your wallpaper. Simply use the -output and the -num_times 1 options on the command line and xplanet will run, save the generated image to the file specified by -output, and exit. Be sure to set the -num_times 1 option, or xplanet won’t close after generating the image.

Next, you’ll want to set your wallpaper to be the saved image. Do this just like you would normally set your wallpaper; you can go to the System menu in either Ubuntu or Fedora, and under Preferences select Appearance.



Or, you can right-click on the desktop and select Change Desktop Background. Either way, click the “Add” button and add the image you saved with Xplanet. Now that will be your background.

But, there’s more to do!

4. Getting the clouds

One of the coolest features of xplanet is that it has the ability to overlay a “cloud map” showing the current cloud coverage on Earth. If you’ll notice in the screenshot at the beginning of this article, my desktop includes a cloud map overlay. There are two choices for setting up the cloud map:

• Download a static cloud map

If you don’t care about having the cloud coverage update itself periodically, you can download a single image and use it whenever xplanet renders the Earth. Simply save this file somewhere, and proceed to the next section about “Creating the configuration.” Or, for more fun…

• Periodically download an updated cloud map automatically

The other option is to setup a script to periodically download an updated cloud map, so that your wallpaper matches the current cloud coverage of the planet. It’s a little tweak, but it’s definitely fun. If you decide it’s worth the effort, here’s what you need to do:

1. Setup the script to download the cloudmap automatically

Download this script and decompress it. This script was written by Michal Pasternak and is in the public domain. It is useful because it checks if a new cloud map is available before actually downloading it, which saves a lot of bandwidth.

2. Run the script once to generate the cloudmap

Once you’ve downloaded the script, make it executable by running the following command in its directory:

chmod +x download_clouds.py

Next, run the script so it will download the initial cloud map:

./download_clouds.py

By default, the script saves the map with a filename of clouds_2048.jpg.

Now continue to the next section, which will show you how to setup the configuration file in order to use the cloud map. Keep reading to the “putting it all together” section to create an automated cron job so the cloud map will update itself periodically.

5. Creating the configuration

If you’re satisfied with the image you generated previously, that’s fine; you can skip this step. But if you want to add clouds or lights, you’ll need to modify the xplanet configuration file to let it know where to find these added goodies. It’s done in a text editor, but it’s pretty simple.

The easiest way to proceed is to modify the default configuration file, by opening it in either vim or your favorite graphical text editor. On both Ubuntu and Fedora, the default configuration file is located at /usr/share/xplanet/config/default. You’ll need to be root to edit it (optionally, you can copy it to your home directory and then edit it, if you don’t want to make system-wide changes).

Inside /usr/share/xplanet/config/default, search for the line [earth], which specifies the options to use when drawing a picture of Earth. Here’s my [earth] section of the configuration file:

[earth]
“Earth”
color={28, 82, 110}
cloud_map=/home/jdeprizi/bin/xplanet/clouds.jpg
night_map=/home/jdeprizi/bin/xplanet/night_xplanet1.jpg
#marker_file=earth
min_radius_for_label=0
#satellite_file=iss

While you should certainly check out what happens if you uncomment satellite_file or marker_file, the two things I want to mention are cloud_map and night_map. The cloud_map option should be set to the image you specified in the section above, “Getting the clouds.” This tells xplanet to overlay the cloud map onto the generated image, and yields a result similar to the screenshot at the beginning of this page.

When you specify the night_map option, xplanet will draw city lights in the part of the Earth where there is no sunlight, like so:



Your night map should be located in /usr/share/xplanet/images with a file name night.jpg. In the above example, I copied it into my home directory.

6. Putting it all together – cronjob and login trigger

Now that you have your background looking just the way you want, you’ll need to configure a scheduled job to make sure that it updates the daylight and the cloudmap automatically. The easiest way to do this is by scheduling a job using cron and gnome-schedule. Use the command you came up with above, and schedule a job to run as often as you want — I generate a new image every 30 minutes. I also download the cloud map every 12 hours:



You’ll also need to have the map generate when you login, especially if your computer is off for long periods of time. If you don’t do this, the map won’t regenerate until the next cronjob kicks off, and you could be looking at a very outdated image for up to an hour. To have the wallpaper refresh when you first login, simply go to the Startup Applications item under System and then Preferences. Add a new item with the same command as you used above, to generate your background:



Now your wallpaper will automatically refresh whenever you login.

7. Now let everyone see!

If you’ve followed through to this point, you should have a fully-functional xplanet wallpaper that will automatically refresh itself, following the interval you specified. Awesome! Show off your work by posting a screenshot to the TechThrob Facebook gallery. Also, leave a comment with the options you use to generate your background, so others can see what kind of cool tweaks you’re using!

First, install the gnome-schedule application. As usual, this is a quick one-liner in most popular distributions:

Ubuntu:sudo apt-get install gnome-schedule
Fedora:sudo yum install gnome-schedule

Once it’s installed, you should see a new “Scheduled Tasks” application in the System Tools menu.

The task scheduler gives you three options for creating a new task. You can chose a recurring task, which is the primary use of gnome-schedule as a frontend for cron. It also allows you to create a task that runs only once; this actually uses the at service.

Additionally, you can create custom schedule templates; for example, if you want to run tasks on certain holidays, you can create a template for that so you don’t have to re-create the cron rules each time. Most users won’t need this feature, but it’s nice to have.

The main window shows you the currently-scheduled tasks:

It’s a bit spartan, and there’s definitely no eye candy, but if you’re at all familiar with cron you’ll be very comfortable with the output.

The primary use of gnome-schedule is to setup cronjobs, and it does it very well. The single-window configuration allows you to specify a name for the job as well as the command to run. The date and time wizard accepts standard cron syntax, but a major feature is a GUI tool that sets up the schedule for you.

For example, by choosing whether you want to run a task every day, every X number of days, or in a range of specific days, you are able to set a schedule with a very high level of precision. This is the precision that cron is known for, and using gnome-schedule you are able to take this power and implement it via a convenient graphical interface.

Gnome-schedule can create one-time jobs, in addition to recurring tasks. Creating a single-instance job actually uses the at daemon rather than cron (advanced users should be familiar with the difference, but from our current perspective it doesn’t matter). When you create a one-time job, you get a simple window to give the job a title and a time to launch, and to say what commands you want to run at the scheduled time.

In the above screenshot, I am creating a scheduled task to pop up a reminder window (using Zenity) at 11:55 to tell me that I need to post something to TechThrob (although I assure you, I don’t need any reminders for that). Of course, you can use this for anything — you might want to schedule an alarm to wake you up in the morning, or a reminder to pick something up on the way home.

I find that the single-instance scheduler is actually incredibly useful, perhaps even more useful that the repeating schedule. I never really used at when it was just a command line utility; I knew it was there, but I didn’t see a need for it. Once I started using gnome-schedule, though, I found that scheduling reminders and automated jobs was very easy and incredibly useful.

The scheduler is incredibly well documented, and you can read the help by clicking the gnome help icon in the main window. I ran into a small problem trying to get graphical applications to launch (zenity, in the above example), but I quickly found the answer in the help (you have to specify the DISPLAY variable). If you’re curious as to how something works, or you’re having some problem, I suggest giving the documentation a look before you get frustrated or ask for help.

Have you ever used cron before? What kind of tasks do you have automated with cronjobs? I’m also curious to know if others find the cron syntax as confusing as I do – I’ve always had to lookup cron syntax when I need to do something more complicated than setting a task to run at a specified minute or hour (hence why I love gnome-schedule’s GUI so much!). Leave your thoughts in the comments below!

Note: You’ll need to be root to run most, if not all, of these commands.

I’ve used iftop for a long time, because it’s a very powerful tool that gives you lots of details about your network connections. Here’s what you need to know about iftop to get started:

1. Basic Usage

iftop is easy to use if you just want to see your current network connections and how much bandwidth is being used by each remote host. Simply launch it from the command line, passing the -i option with the interface you want to monitor, and optionally the -B option to display values in bytes (the default is to display in bits). So for example, to monitor the wlan0 device in bytes, you would run:

iftop -i wlan0 -B

There’s a lot of information displayed on the screen, but it is formatted intelligently and quickly becomes easy to digest. Here is what the basic screen is showing you, when you first launch iftop:



As you can see, the display is packed full of useful information (click on the image above for a larger version). Don’t worry if it’s a little overwhelming at first; it won’t take long to get used to reading the display, and you’ll appreciate having all this information available so quickly.

Now let’s move on to some of the more powerful features that iftop provides.

2. Filtering networks, hosts, and ports

While it’s nice to see all the hosts your computer is talking to, it’s often the case that you’re only interested in a certain segment of the network. iftop allows you to filter connections by network, host, and port, which gives you complete control over which connections are displayed.

iftop accepts pcap-filter formatted filters on the commandline with the -f flag. Below is a table of some of the filers you might want to use with iftop:

dst host host	src host host
dst net net	src net net
dst port port	src port port
dst portrange start-end	src portrange start-end
gateway gateway
ip proto protocol

For example, to view only traffic going from your local machine to google.com over eth0, you could run:

iftop -i eth0 -f “dst host google.com”

Or to see only ssh traffic over wlan0:

iftop -i wlan0 -f “dst port 22″

Additionally, iftop allows you to set arbitrary filters based upon regular expressions. It’s important to note that when you specify a filter with a regular expression, you are only filtering the on-screen output, whereas using a pcap filter (above) will filter what iftop actually listens to. As a result, the totals displayed at the bottom of the screen won’t be affected by regex filters. You can press the l key to enter regular expression filters while iftop is running.

3. Controlling the interface and the online help

Once you’ve got the information you want on the screen, you’ll need to be able to move around and tweak the exact output. There are many options, and the easiest thing to do is just hit the h or ? key to see the on-screen help:



Notice that you can toggle things like hostname and port resolution, port display, and whether iftop sorts by destination or source. Showing connections by port is useful for monitoring throughput on programs that create many connections, while turning DNS resolution on might make it easier to read the display if you are watching things like web traffic.

Now I’ll show you how to use nethogs, which is a great little program that was mentioned by one of TechThrob’s readers (thanks, dasen!). Whereas iftop displays network usage by destination IP address and port number, nethogs takes a process-oriented approach and shows you usage based on the program that is accessing the network.

nethogs is much simpler than iftop and doesn’t have as many options. You can specify the interface to listen on when you launch it:

Which will give you a screen similar to the following, showing the processes that are sending or receiving traffic on that interface.

While it’s running, you can use the m key to toggle between units (megabytes, kilobytes, and bytes) and to change whether you are viewing instantaneous throughput or the total throughput since nethogs was started.

Being so simple is the greatest advantage of nethogs, since it makes it much more user-friendly than iftop and lets you see in an instant what applications are using your network, and how much bandwidth they are using. If you think your network is being hammered and you want to know who to hold responsible, nethogs is probably the command you want.

With great power comes great responsibility. If you have root on a machine that is routing a lot of traffic, tools like this will let you see what people are doing on the network; don’t be evil. Use these tools to monitor your own traffic, and to troubleshoot problems, but don’t spy on other people.

1. iftop

As you might be able to deduce from the name, iftop is a top-like application for network interfaces. You need to be root to run it, and it can be installed using the command:

Ubuntu: sudo apt-get install iftop
Fedora: sudo yum install iftop

Run it in a terminal by specificying the interface you want to use, and optionally a -B (make sure you use a capital B) to display values in bytes rather than bits:

# iftop -i eth3 -B

The result is a screen showing you all the current connections on that port, as well as the throughput for each connection. Like top, iftop is full of features; press the h key to get an online help showing the options.



Here you can see my network throughput as I run a speed test on dslreports.com. You’ll notice that there are separate indicators for incoming and outgoing traffic, and that three values are given on the right-hand side; these are the 2, 10, and 40 second averages.

If you find iftop useful, you might want to read more about monitoring network traffic.

2. iotop

While iftop is great for monitoring network throughput, iotop is the utility you want if you need to measure disk access. It’s easily installable from the commandline:

Ubuntu: sudo apt-get install iotop
Fedora: sudo yum install iotop

And you can run it with a simple sudo iotop command. What you’ll get is a listing of processes, like in top, but with their disk access information:



In the above screenshot, I’m running a simple dd test to write to my local hard disk, which you can see as the first item in the window.

iotop will tell you how much data each process is reading and writing, as well as if a process is using swap (which can be really helpful if you need to determine which process is thrashing your disk).

3. pv

The pv command isn’t a top-like application at all, but I’ve found it incredibly useful and I wanted to share it with you here. The pv utility is used between pipes on a command line and measures the speed of data throwing through that pipe. For example, here I am using pv to measure how quickly data is flowing between two dd commands:



Installing pv is a simple one-liner:

Ubuntu: sudo apt-get install pv
Fedora: sudo yum install pv

This is a great way to see how quickly one command is passing data to another command, when neither command is giving any output to report on its progress. If you’ve ever setup a pipe on the command line and wondered if it was actually doing anything, pv is the command you’ve been looking for. You can use it not only when you need to know the speed of data moving through a pipe, but whether there is any data moving at all.

4. latencytop

I wanted to give a quick mention to latencytop since I’m sure someone will ask about it if I don’t mention it. latencytop is a tool that shows you what resources a given process (or processes) are waiting for on your system. For example, it can tell you that a program is waiting on disk I/O or waiting to get a file lock. latencytop is a great tool, but on the whole it’s more useful for developers than end-users. If you’re interested, this is what it looks like:



And this is how to install it:

Ubuntu: sudo apt-get install latencytop
Fedora: sudo yum install latencytop

Did I miss your favorite tool? What are your go-to commands when your system performance isn’t what it should be? Leave your helpful tips in the comment section below!

1. Software Center enhancements

A major focus of Ubuntu 10.10 is improving the software center, addressing many of the usability problems that have been sources of complaints in the past. Among these changes are:

• Better Search

I’ve heard this complaint quite a bit, including in the comments of my article covering things new Linux users need to know. I had posted a screenshot of an application there, and someone tried to find it in the software center by searching for “Disk Analyzer”, which is what’s displayed in the title bar of the program. Unfortunately, no results were returned. In the next version of Ubuntu, the Software Center will show suggested results when you search for something that gets no hits.

• Improved dependency display

Most users, even the more advanced members of the crowd, don’t particularly care about the package dependencies of a particular program. Managing that kind of information is trusted to the package management program, and there’s really no reason for a user to be presented with that data unless she asks. In the next version of the software manager, everything but the application itself will be hidden, with the option to show “Technical items” only when the user specifically requests it.

• Add-on packages and media

Many packages offer “add-ons” which extend the feature-set or usability of the program. A good example of this is the Firefox browser, which today has several add-ons available in the Software Center, such as the Ubuntu extension. In the next version of the Software Center, add-ons will be much better organized.

2. OneConf: Sync your configuration between machines

OneConf will allow users to share their Ubuntu configurations between multiple machines. Realizing that people work on more than one computer, and taking a cue from browser sync features, OneConf will allow you to store your installed application list and those applications’ settings to the UbuntuOne service. You’ll then be able to migrate this list to another machine, or to use it as a configuration restore. It will support multiple configuration specifications, allowing you to keep separate lists for different types of machines (home vs. work; desktop vs. netbook, etc..).

3. Post-Release application delivery

Developers and users alike will look forward to the ability for new packages to be introduced to the distribution after it has been released. Although the process is not finalized, there will be a process by which developers can submit their packages for review and inclusion into the software repositories, even after a major release. This means that Ubuntu users will be able to receive new packages without upgrading or manually seeking them out, which is the case today.

4. Chromium as the default netbook browser

Ubuntu 10.10 aims to improve netbook support (using its Ubuntu Netbook Edition release), and part of this is a migration to the light-weight Chromium browser. Many Linux users are already familiar with Chromium, or its close Google-branded relative, Chrome, as a speedy alternative to Firefox.

5. Better touchscreen support

Touchscreen support is another area where 10.10 should show significant improvements. On the drawing-board for this release is to improve existing applications’ touch-friendlyness by tweaking GTK, icon settings, and other theme options. Additionally, support for gestures in Compiz may also be included. This is an area where you can expect to see improvements beyond the immediate future, as touchscreens become more common and Ubuntu moves to support this market. In the future, we’ll likely see further enhancements, such as the inclusion of a built-in on-screen keyboard.

Do any of these features make you excited for Ubuntu 10.10? If so, which ones? And if not, what would you want to see in Ubuntu 10.10? Leave your opinions in the comment section below.

References
1. Software Center Improvement Speclist
2. OneConf Spec
3. PostReleaseApps Process Wiki
4. Chromium as Default UNE Browser
5. Touchscreen Improvement

1. It puts you in charge

Between my day job at a software company and my work on TechThrob, I spend a lot of time on the computer, doing technical things. Nothing compares to the power of Linux when it comes to managing technology, whether it be administering severs, writing code, or interfacing with computers in some other way.

Windows intentionally puts a barrier between the user and the underlying machine, whereas Linux completely exposes every aspect of the system: all you have to do is open up a terminal. As someone who works with technology on a daily basis, this is invaluable to me.

2. Software repositories

I don’t have time to manually search the Internet for the right tool to do a job. When I need to download a new application and I don’t know which program is appropriate, it’s usually because I’ve hit some stumbling block that’s preventing me from moving on with my work. The last thing I want to do is to start googling for reviews of shareware programs that may or may not work, and might even be harmful to my computer.

Having a centralized software repository completely solves this problem. Instead of spending time searching the web for the appropriate application, I just execute a single “yum search” or “apt-cache search” command, and find precisely what I need in seconds. It’s hard to appreciate just how priceless that can be until you are in a situation where you need to solve a problem immediately and you don’t have a software repository available.

3. Safety

A secondary advantage of centralized software repositories is that they are safe. Even as someone who understands the dangerous parts of computing (trojans, worms, malware, etc..), I know that even the most well-educated, best-protected Windows user is still vulnerable due to the sheer number of exploits available. When you’re using an operating system that’s under constant attack by virus writers and bot-net operators, you’re eventually going to get stung, no matter how well you protect yourself.

Using Linux alleviates most of this concern. It’s important not to develop a feeling of invulnerability, because you still need to be careful (for example, don’t run scripts or programs from untrusted sources, or you’re just asking to get bitten) but there isn’t the ever-present danger that exists when you are using a Windows machine.

4. It’s got the apps

I have to admit that what I consider a “killer app” isn’t necessarily the same as what a graphics artist or a sound engineer would consider a “killer app.” But as a power user, Linux gives me what I need to make my life easier and more productive. From utilities that track down IP address conflicts to programs that securely wipe my hard drive, the toolset that you get when you run Linux is simply unrivaled by a standard Windows installation.

Admittedly, Windows may very well have similar programs available for it, but that simply goes back to my second point about software repositories– on Windows, I’d have to hunt these applications down one by one, but on Linux, everything is a single command away.

5. It’s rock-solid

I don’t have the time to be rebooting my computer because it’s grown inexplicably slow. I can’t be bothered trying to figure out why some application won’t close, and I certainly don’t feel like dealing with the frustration of having a single program lock up my entire machine. I need my computer to work for me, not against me, and this is where Linux truly excels.

Windows tries to hold your hand when you use the computer; consequently, when it trips and falls, you wind up face-first in the mud along with it. Linux doesn’t do this — if a process misbehaves, it rarely affects anything else on the system. If it’s eating up too much CPU or memory, I can just kill it; on Windows, a run-away program might lock-up the machine to the point of requiring a reboot.

6. It’s what I know best

I’m only counting this as half a reason, but it’s important to point out that while a lot of the above applies to Linux because of the way it’s written and the way it works, I am also able to take advantage of Linux because I’ve had years of experience with it.

I know a handful of Windows power-users who are able to do some pretty impressive things with that OS because they’ve learned the best way to use and manipulate it. But for me, it’s much easier, faster, and better to use Linux — precisely because I don’t need to manipulate it into doing what I need.

Do you consider yourself a power user? If so, what are some tools that are indispensable to you, and could you be as productive if you weren’t running Linux? Share your favorite things about Linux in the comments below!

1. Install the Nautilus Actions Configuration utility

First, install the nautilus-actions package, which provides a graphical utility for editing the Nautilus right-click menu. We’ll use this utility to add the secure delete option. You can install the nautilus-actions package by using the following command:

Fedora	Ubuntu
sudo yum install nautilus-actions	sudo apt-get install nautilus-actions

Once installed, you can configure the new menu option.

2. Configure the new menu option

Now you should have an option in the Preferences menu called “Nautilus Actions Configuration.” This will allow you to specify the new option for securely deleting files from the file manager. Click the “define new action” button (or select it from the file menu). Specify the information as shown below (feel free to use whatever text you want as the label and tooltip; this is up to you!):



Next, click the “Command” tab. This is where you’ll specify what the new menu item will actually do when it’s clicked. As you see below, I am using the shred program as my secure-delete method, which is located at /usr/bin/shred. For the paramters, I need to specify -u in order to have shred delete the file once it’s been overwritten.



Clicking the “legend” button shows that the option to pass a list of files to the commandline is %M. So my full paramters string becomes “-u %M”. You should feel free to customize the command however you like, by reading the manpage for the command you are using, and by referencing the legend to figure out which substitution strings to use. For example, you might want to use the “-z” option for shred, which will use a final pass of zeros, to “hide” that the file has been shredded, or the “-n” option to specify the number of passes to use.

Finally, you’ll want to set the conditions under which the menu item appears. In the “Conditions” tab, I’ve selected for the secure delete option to appear only when files are selected, and to allow it to be used on multiple files at once. Here is how my Conditions tab looks:



With that, double-check that you’ve entered everything correctly, and click “Save” in the file menu.

3. Test it out

That’s all you need to do — the only thing left is to test that it works. Find some files you want to shred, or create a few dummy files (obviously don’t use this on data you need to keep!). I used a bash “for” loop to create a few test files, and then opened Nautilus to the directory where I put them.



Selecting them all and right-clicking, I see the secure deletion option. Clicking it and watching in “top” shows that it is working correctly. Those files are gone!

4. You can use this for anything!

The nautilus-actions utility can be used for more than just adding secure-deletion options to the file manager; you can use it to execute any command-line operation you want. What are some things that you might add to the Nautilus menu? What other file manager hacks do you find useful? Leave your ideas in the comments below!

8SQHVUP4GCZ8




]]> http://techthrob.com/2010/07/07/adding-a-secure-delete-option-to-nautilus-file-manager-in-linux/feed/ 17