From their home page:

BleachBit quickly frees disk space, removes hidden junk, and easily guards your privacy. Erase cache, delete cookies, clear Internet history, remove unused localizations, shred logs, and delete temporary files. Designed for Linux and Windows systems, it wipes clean 70 applications including Firefox, Internet Explorer, Flash, Google Chrome, Opera, Safari, Adobe Reader, APT, and more.

It is available for most Linux distributions. Here is a great write-up on it from Linux Magazine.

If any of the output runs off your screen, just add |more to the end of any of these commands to see the output one screen at a time and hit the spacebar to go to the next screen, or Q to quit.

Processor type:
$ cat /proc/cpuinfo

Is the processor using 32 or 64 bit instruction set:
$ cat /proc/cpuinfo | grep flags | grep lm
If you get some output you have a 64 bit CPU. If you receive no output, then you’re using a 32 or even 16 bit CPU. The reason this is the case is that the CPU yields many flags that tell Linux what sort of processor it is, and the lm flag informs Linux that the CPU is a 64 bit processor. Grep as a command filters output. Feel free to run this command without the grep suffixes (cat /proc/cpuinfo) to see the full output of your CPU details.

What hardware (audio, video, disk controllers, etc) is in my Linux box:
$ lspci -tv
(The -t switch groups similar devices together for easy reading and -v offers more verbosity.)

To easily filter out the above command to just show graphic card information:
$ lspci | grep VGA

What USB devices are plugged in:
$ lsusb

Check the size of the hard drive and what hard drives are available in the system.
This command will also list USB drives and sticks. You need a root permissions to execute the fdisk command:
$ sudo fdisk -l | grep GB

Show info about a particular hard disk including firmware revision (replace sda with the appropriate drive as listed from the above command):
Note: This will only work on internal disks, NOT USB drives.
$ sudo hdparm -i /dev/sda

Check what partitions and file system is in use on my hard drives (same as the above command, but essentially more verbose):
$ sudo fdisk -l

Locate CD/DVD-ROM device file which offers a CD/DVD-ROM’s make and model info:
$ wodim –devices
or
$ wodim –scanbus
The above command will scan your entire system bus for attached devices (this won’t include USB Devices as they are not direct-bus-attached devices).

What modules are currently loaded:
$ lsmod

get a information about any particular module:
$ modinfo module_name

remove modules:
$ modprobe –remove module_name

load a modules to the kernel:
$ modprobe module_name

What hardware is using which module.
The -v switch is for vebosity, where -vvv is EXTRA verbosity.
$ lspci -v
or
$ lspci -vvv

Check for PCMCIA cards:
$ lspcmcia

How much RAM is installed in my Linux and how much of it is in use (megabytes).
It will also include swap memory:
$ free -m
There is a gigabyte switch, but it *rounds* it down, so it isn’t very accurate for RAM info:
$ free -g

Check sound card settings. This command will reveal whether your sound card is installed and what modules are in use:
$ cat /dev/sndstat

Available wireless cards:
$ iwconfig

What speed is set to FANs:
$ cat /proc/acpi/ibm/fan
If this command doesn’t work, then feel free to peruse the /proc/acpi directory on your system. You will find info available on your CPU, AC Adapter, Battery, etc. Some info is available here, and your mileage may vary for viewing any of the files in /proc/acpi.

Get a battery information on your laptop (assuming it’s been installed):
$ powersave -b

To find out what Linux Kernel you’r running:
$ uname -a

To find out what distribution of Linux you’re running:
Run any of these commands, as depending on your distribution some may or may not work.
$ cat /etc/issue
$ cat /proc/version
$ dmesg | head -1

Get a recent history of system reboots:
$ last reboot

To open any file from command line using the default application (will launch the correct graphical application for the file, as though you had doubled-clicked the file graphically):
$ xdg-open ./filename

To monitor all active network connections, and update live every second:
$ watch -n.1 ‘netstat -tup’

To passively list all connections, active or inactive:
$ ‘netstat -tupl

Not every EEE PC is built the same, nor do they use the same video drivers. Therefore first, you have to find what the minimum and maximum resolutions are available for your unit. To see this, simply type this in command line:

xrandr

As an aside, the commands below work with any Linux system, EEE or otherwise. This command will tell you the minimum and maximum resolution of your screen plus the available “standard” resolutions.. mine were:

Screen 0 (the local screen): min 320×200 ………. current 1024×600 ……… max 1024.x1024

Available standard modes were: 1024×600, 800×600 and 640×480.

To max out my screen resolution I chose to work with 1024×1024, that would give me the most pixels (+ the most panning) but it would show me more of an application in a glance. Normally working in 1024×600 (native resolution of the EEE PC 1000HE) I’d have to hit F11 to go to full-screen mode in Firefox. Now, running in 1024×1024, I can run without full screen mode and can see a nice portion of my browser window. The panning isn’t much of a bother, and I feel less constrained on the desktop.

To change the resolution to the max specified by running xrandr above, type this:

xrandr --output LVDS --panning 1024x1024

. . . and voila !  Instant pan/scan screen. To change it back to the original resolution …

xrandr --output LVDS --panning 1024x600

Place these two commands in separate scripts with execute rights, then attach to a launcher and place on your taskbar and/or desktop and you can do this in one click.

###

UPDATE: On Ubuntu 9.10, this command WORKS, however, the local display is now LVDS1, not LVDS.

They’re sorted newest-first, so click the <previous entries> at the bottom of the page to see the prior scripts.

Here’s also a few links on learning the basics of shell scripting:

IBM’s extensive site on the subject.

Linuxcommand.org

Simply sudo apt-get install cpulimit. More info can be found here.

I recently had to run a ReiserFsck myself on one of my primary home Linux boxes off a Live-CD; worked very well!

Here’s the link for reference:

NOTE: The above link is just a broad stroke introduction to some of the more commonly used repair functions, a lot more research is needed by the user if in-depth recovery of a hard disk or repair is required.

Don’t forget to chmod +x ./script-name on each.

Script 1

#!/bin/bash
#
# Dump m4a to wav (first step in conversion)
for i in *.m4a
do
mplayer -ao pcm "$i" -aofile "$i.wav"
done

Script 2

#!/bin/bash
#
#Second step... use lame to convert into .mp3

for i in *.wav
do
lame -h -b 192 "$i" "$i.mp3"
done

Script 3

#!/bin/bash
#All the m4a's are now mp3's, however..
#the file will look like "filename.m4a.wav.mp3"
#So, to clean that up we remove extraneous extensions.

for i in *.mp3
do
x=`echo "$i"|sed -e 's/m4a.wav.mp3/mp3/'`
mv "$i" "$x"
done

python -m SimpleHTTPServer &

This will launch a very simple web server in the directory from which the command was run, and will establish that directory as root. You can then access it from your local machine by typing in this URL in your browser: http://127.0.0.1:8000/

By default, it starts the web server on port 8000 but that can be changed using:

python -m SimpleHTTPServer 9090

which starts the server on port 9090, and can be accessed via the browser or any HTTP client using:

http://127.0.0.1:9090/

If you want to access these files from outside your home network, I’d recommend SSH with a local port redirect instead of raw port forwarding, which I do talk in a roundabout way here.

For any that are interested, this is my .screenrc

hardstatus alwayslastline
hardstatus string '%{= kG}[ %{G}%H %{g}][%= %{=kw}%?%-Lw%?%{r}(%{W}%n*%f%t%?(%u)%?%{r})%{w}%?%+Lw%?%?%= %{g}][%{B}%Y-%m-%d %{W}%c %{g}]'
#
# Default screens
# syntax: screen -t NameOfScreen ScreenNumber ShellCommand
screen -t "Work SSH" 0 /home/name/workssh
screen -t "Home 1" 1
screen -t "Home 2" 2
screen -t "Home 3" 3
screen -t "Home 4" 4

The hardstatus string gives me the local computer’s hostname on the left, lists the shells themselves in the middle, and on the right offers the date and time.

Line 6 with “Work SSH” executes the shell script “workssh”, labels the window “Work SSH” and assigns it to virtual shell number 0. This shell script allows me to SSH to one of my work servers offering me a series of local listening ports that forward to remote IP’s and ports.

Shells 1-4 are local bash shells.

Assuming you have VirtualBox remote port forwarding engaged (which will forward ports into the VM, for example if you’re running Windows XP as a VirtuaBox VM and you want to RDP into it on port 3389), I discuss port forwarding on VirtualBox hosts, here. Though VirtualBox supports a quick and easy VRDP option to enable for easy RDPing into a box without having to do any of the complex port forwarding, my port-forwarding post explains how to forward other ports to your VM for any other applications or if you want to RDP into your Windows VM on a port other than 3389.

To engage the VM over an SSH session, simply type:

VBoxHeadless -startvm "Name of VM Here"

You can also force VirtualBox’s VRDP (virtual RDP) via command line by adding the switch “–vrdp on”, though it’s usually on by default.

If you’re not sure of the name of the VM, simply type

VBoxManage list vms

Once running, you can simply run …

rdesktop localhost:3389

Assuming you have a -L 3389:ip-address-of-VirtualBox-host:3389 in your original SSH command, and you can now RDP into your Windows XP box which would be running in a VM over Linux.

Source: HowToForge

Solution 1

We are going to use qemu-img tool to this.QEMU disk image utility

First Install qemu  qemu-img is included with qemu package using the following command

sudo apt-get install qemu

Convert a VMWare Image to VirtualBox Image

Convert VMWare image called whatever.vmdk to /tmp/debian.bin (or a path of your choice).

qemu-img convert debian.vmdk /tmp/debian.bin

Now use VBoxManage to get back image in native format:

VBoxManage convertdd /tmp/debian.bin debian.vdi

Solution 2

VirtualBox can run VMs created by VMware Workstation or Server for this you need to import vmdk files using the following procedure

• Start Virtual Box

• Goto File > Virtual Disk Manager

• Click Add.  Locate and select the copied .vmdk file. Click OK.

• Create a New VM as usual using the added vmdk file

• Boot the VM

lsof +D /media/disk1

This will return the command and process ID of any tasks currently accessing the /media/disk1 directory. You can then locate them, or use the kill command to finish them off.

Bonding is creation of a single bonded interface by combining 2 or more ethernet interfaces. This helps in high availability and performance improvement.

How to setup dual-dual bonding (two bonds of two interfaces each) on Ubuntu as quickly as possible.

1. Add two lines to /etc/modules

bonding bond0 -o bond0 mode=1 miimon=100

bonding bond1 -o bond1 mode=1 miimon=100

If you’re very good at managing your time, just remember that miimon’s option determines how often the bond is monitored for failure and that mode can be one of:

0 – Round robin balancing

1 – Active back-up

2 – Transmit based on MAC address for load balancing/fault tolerance

3 – Broadcasting – provides fault tolerance by transmitting on all slave interfaces

4 – Aggregates links, assuming all nics support same speeds and duplex settings

5 – Transmit load balancing – balancing is handled by the bond based on load

6 – Same as 5, but also uses arp to balance load “better

2. Install the ifenslave package if you haven’t already. You can use apt-get to grab it if you don’t:

sudo apt-get install ifenslave-x.x

3. Ensure that the package actually installed:

sudo dpkg –get-selections | grep enslave

ifenslave-x.x install

4. Set up your interface files:

# cat /etc/network/interfaces (only including the parts you probably need – substitute IP addresses, netmasks, etc):

auto lo

iface lo inet loopback

auto bond0

iface bond0 inet static

address 10.10.125.88

netmask 255.255.255.0

network 10.10.125.0

gateway 10.10.125.1

post-up ifenslave bond0 eth0 eth2

pre-down ifenslave -d bond0 eth0 eth2

auto bond1

iface bond1 inet static

address 10.10.127.88

netmask 255.255.255.0

network 10.10.127.0

gateway 10.10.127.1

post-up ifenslave bond1 eth1 eth3

pre-down ifenslave -d bond1 eth1 eth3

5. Add lines to the bottom of your architecture’s modprobe files, reboot
and pray:

sudo cat /etc/modprobe.d/arch/i386

alias bond0 bonding

options bond0 mode=1 miimon=5000 max_bonds=2

alias bond1 bonding

options bond1 mode=1 miimon=5000 max_bonds=2

Most people think you need to kill your SSH session to add a new -L option, then reinitiate the SSH session; this is NOT true.

You can open an internal SSH shell within an SSH session to add new redirects!

From within an SSH session, simply type:

~# then hit <enter>

~C then hit <enter>

Then type: help and <enter>

You’ll see a listing of available commands. To add a new local redirect, just type

-L 4567:192.168.0.12:5900, then hit <enter>

…and voilà, you’ve added a new local redirect. Just hit <enter> once, and you’ll be dropped back into command line.

There are other escape-commands. Just type ~? from within an SSH session for more escape-commands.

FYI: This also works for remote redirects, as well.

Here’s a Google search link offering more info:

Here’s a link to Clonezilla SE (Server Edition).

Ext4 which has now been declared stable, and no longer experimental, will be the successor to the long-standing ext3. It can support a file system up to 1 exabyte in size (that’s 1,048,576 terabytes to everyone else) and single files up to 16TiB.

Ext4 is likely to become the defacto standard file system in most Linux systems, perhaps except the most mission-critical of servers whose admins will likely opt for a file system longer in the tooth (like ext3 or XFS). Ext4 will likely be the file system to go head to head with SUN’s ZFS.

Kernel 2.6.28 will also support laptop disk shock protection, and the minstrel WIFI rate control algorithm.

More details on the other improvements here, and here.

Some more details here, which are easier to digest.

Here’s a video of it:

Here’s the blogpost on it:

It is a complete solution for every organization to create, collaborate, share and publish all its internal and external documents.

You and your team can create and collaborate on:

• Text documents
• Spreadsheets (coming soon)
• Presentations
• Task Lists
• E-mails
• Calendars
• Web Links

All it requires is an xampp server (essentially a Linux server). I am going to test this with some colleagues and see how viable it is for a production environment, but on the surface it looks to be worth evaluation.

You can try a demo here. Once the spreadsheet function is available, this will become much more interesting.

The Linux Basement also hosts a podcast worth checking out which centers around web hosting and network management, but is overall an informative show. They do about 2 episodes per month.

They also have a fairly active user submitted article area, as well as a forum.

One of the most common mistakes is not checking log files (reason #10). I often find it hard to keep up and sometimes parse the variety of system logs on a system. A great tool for managing this is called LogWatch. While not a GUI, it summarizes system logs into a convenient report that you can cron to run regularly and email to yourself, or just monitor. It’s highly configurable with varying degrees of detail.

From the site:

Logwatch is a customizable log analysis system. Logwatch parses through your system’s logs for a given period of time and creates a report analyzing areas that you specify, in as much detail as you require.

GroundWork Monitor Community Edition provides a essential IT monitoring solution that enables you to maintain network visibility and control.

This is a single server deployment that leverages the strengths of 15 other open source projects such as Nagios, rrdtool and nmap. It combines the results of these projects in to a system that evaluates the status, events and performance of monitored devices and presents these results in an easy to understand web application.

This open source project distributed on sourceforge.net is available under the GNU Public License (GPL v2). It is ideal for operations who already have deployed one or more open source monitoring projects and need to expand its capability.

Run these 3 commands from shell. SUDO not required.

VBoxManage setextradata “name of vm” “VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/HostPort” 2222

VBoxManage setextradata “name of vm” “VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/GuestPort” 22

VBoxManage setextradata “name of vm” “VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/Protocol” TCP

Where…

• “name of vm” is your virtual machine name
• ssh is the name of the service. This is _required_ to be unique among the three commands.
• Ports 2222, 22, TCP respectively with your desired host port, guest port, and protocol.

…then shutdown your VM and Virtualbox host application, and restart them.

To Confirm the above settings are in place:

VBoxManage getextradata “name of vm” enumerate

To remove the settings made above from the VM: (again substituting the variables as appropriate).

VBoxManage setextradata “name of vm” “VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/HostPort”

VBoxManage setextradata “name of vm” “VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/GuestPort”

VBoxManage setextradata “name of vm” “VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/Protocol”

Click here for details:

Another way to do this is to use CloneZilla.

1) Dell: Dell is currently selling four laptop systems starting from $549 to $1049; the linux desktop system available starts at $448. All these system are pre-installed with Ubuntu 8.04. here:

2) system76: They are Mark Shuttleworth’s personal favorite Linux pre-installed reseller. Like Dell they only sell pre-installed Ubuntu Linux. But they have a wider choice in systems and hardware configurations. Their cheapest laptop starts at $869. More information here.

3) The Linux Laptop Company: As the name suggests they are specialized in selling pre-installed linux laptops; like the previous two vendors their distro of choice is Ubuntu 8.04. They currently sell four models of laptops and the price starts at $699 to all the way up to $1299. More information here.

4) ASUS EeePC: Based on Xandros Linux Asus EeePC is perhaps the most widely used pre-installed linux subnotebook ever sold. There are several resellers for ASUS EeePC, but my favorite is newegg; where the listed price for 1000H 40G is $669. You can get more information about the system here.

5) Linux Emporium: Based on UK, they are Thinkpad resellers with three choices of Linux distro. Their systems can be configured for Ubuntu, Suse or Fedora. Prices start at £376. Get more information here.

6) Linux Certified: They sell a wide variety of laptops with six choices of pre-installed Linux Distros. Fedora 8, Ubuntu 8.04, open SUSE 11, RHEL 5, CenOS 5, Oracle Linux. Their LC2000 series laptops start $699. here.

Source: LinuxHaxor.

your mileage may vary:

_______________

Harder:

- Anatomy of Security-Enhanced Linux (SELinux)

- Understanding AppArmor

- grsecurity

- What is Linux Security Module?

- 10 Basic Linux Security Tips to Implement

- The Best Linux Security Tools

- Linux Audit and Intrusion Detection

- The Unix security audit and intrusion detection tool

- Top 100 Network Security Tools

- The short life and hard times of a Linux virus

- LIDS Secure Linux System

- 10 ways to secure your Linux desktop

Better:

- Beautiful Conky Setup

- Building a Better (Linux) GPU Benchmark

- Last-Exit: A Better Linux Client for LastFM

- The best Linux system repair distribution

- A Better Introduction to Linux User Interface

- Better Linux Dock

Faster:

- How to make Ubuntu extremely fast

- Booting Linux in five seconds

- Improving boot time on a general Linux distribution

- Reboot Linux faster using kexec

- Make Linux Faster and Smoother

- A better, faster desktop search for Linux

- 10 ways to make Linux boot faster

- Fast file copy in Linux

- How fast does Arch Linux install? (video)

Unless you have fully embraced the virtualization movement, you probably have more than one machine in your home or office, particularly if you run more than one operating system, and you probably have more than one keyboard and mouse on your desk. If you would like to regain some desk space without having to purchase a KVM switch, x2x may be the solution. Simply put, x2x takes advantage of the X Window System’s ability to run a display over a TCP/IP network. In this case, you are not running a display as much as you are taking charge of another display’s mouse and keyboard.

More can be found here.

Here’s the full howto.

Howto use Cryptsetup with LUKS support. (Debian Linux).

Linux Command Reference

Linux Security Quick Reference

The Gimp Hotkey Reference

There is a simple applet (reached via the menu “System | Preferences | Keyboard Shortcuts”) which allows you to specify which keys should be used for common activities.

Using the applet you can, for example, setup the keybinding Ctrl-Alt-t to run a terminal. However the list of actions is static, and it isn’t possible to add a new command there.

If you wish to bind Ctrl-Alt-e to launching emacs you’re out of luck – to do this you must use the gconf-editor.

Launch the editor by opening a terminal and running:

skx@pies:~$ gconf-editor

Using the gconf-editor we will need to do two things:

• Define a new command.
• Define the shortcut which should be used to execute that command.

So first of all navigate to “apps | metacity | keybinding_commands“. Once there in the righthand pane select “command_1″ and change the value to /usr/bin/emacs.

Now we need to define the shortcut which will execute it, we do that by visiting the key “apps | metacity | global_keybindings“. Choose “run_command_1″ and set the value to be “<Control><Alt>e”.

Now you should find that pressing Ctrl+Alt+e will launch emacs.

I hope in the future the keyboard shortcut application will allow you to define arbitrary commands and their associated shortcuts.

Preparing for backup

• Depending on why you’re backing up, you might want to:

  • Delete all your emails
  • Clear your browser search history
  • Wipe your saved browser personal details

    If you are not worried about the security concerns, this step is not necessary. Many users explicitly want backups of their email and browser settings.

• Unmount any external media devices, and remove any CDs/DVDs not needed for the backup process.
  • This will lessen the amount of exclusions you need to type later in the process.

Backing up

• Some directories require root or superuser permissions to successfully backup. Gain superuser access by opening a terminal and entering:

  sudo -s -H

• Go to the root of your file system:

  cd /

  We use the file system root in our example, but you use any target destination you want. You can use remote or removable drives as your backup destination.

• Create a backup of your system:

  tar -cvpzf /backup.tgz –exclude=/proc –exclude=/lost+found –exclude=/backup.tgz –exclude=/mnt –exclude=/sys /

• Now, lets explain this a little bit:

  • ‘tar’ is the program used to do a backup
  • c – create a new backup archive
  • v – verbose mode, tar will print what it’s doing to the screen
  • z – compress the backup file with ‘gzip’ to make it smaller

  • f <filename> – specifies where to store the backup, /backup.tgz is the file used in this example

  • Now come the directories we want to exclude. We don’t want to backup everything since some directories aren’t very useful to include.
    • Make sure you don’t include the file itself, or else you’ll get weird results.

    • Don’t include the /mnt folder if you have other partitions mounted there.

      • If you have Partitions in /mnt that require backup, you will need to exclude the folders you do not want backed up.

        –exclude=/mnt/<unwanted_partition>

    • Make sure you don’t have anything mounted in /media.

      • Remove CDs/DVDs and removable media that you don’t need backed up. You can selectively exclude directories in /media if you want removable devices backed up.

  • After all of the options is the directory we want to backup. Since we want to backup everything we use / for the root directory.

• If you want to exclude all other file systems you can use the –one-file-system option in addition to or instead of –exclude.

  • With the –one-file-system option, only the “local” file system is backed up.

    The “local” file system is the file system you have specified, not the file systems mounted under it in the file hierarchy. Use df to see which file systems you have mounted.

    

    tar -cvzf /backup.tgz –one-file-system –exclude=/lost+found –exclude=/backup.tgz /

• Relax while Tar creates a backup of your system. This make take awhile depending on the amount of data that is being backed up and the speed of your processor. When the process is complete you will have a file named backup.tgz in the root directory of your file system. This file may be burned to a CD/DVD, moved to another partition/drive, or even stored on another machine.

• Files that are bigger than 2GB are not supported by some implementations of ISO9660 and may not be restorable. So don’t simply burn a DVD with a huge .iso file on it. Split it up using the command split or use a different way to get it onto the DVD. See man split for further information on split.

  • A possible workaround is the following:

    sudo tar –create –bzip2 –exclude /tmp –one-file-system –sparse / | growisofs -use-the-force-luke -Z /dev/hda=/proc/self/fd/0

    • Note that this only backs up one file system. You might want to use –exclude instead of –one-file-system to filter out the stuff you don’t want backed up. This assumes your DVD drive is /dev/hda. This will not create a mountable DVD. To restore it you will reference the device file:

      sudo tar –extract –bzip2 –file /dev/hda

      At the end of the process you might get a message along the lines of ‘tar: Error exit delayed from previous errors’ or something, but in most cases you can just ignore that.

    • Another workaround would be to Bzip2 to compress your backup. Bzip2 provides a higher compression ratio at the expense of speed. If compression is important to you, just substitute the z in the command with j, and change the file name to backup.tar.bz2. That would make the command look like this:

      tar -cvpjf /backup.tar.bz2 –exclude=/proc –exclude=/lost+found –exclude=/backup.tar.bz2 –exclude=/mnt –exclude=/sys /

Backup over a network

• It is possible to use netcat to transfer the backup between computers.

  • On the receiving end you’ll setup netcat to write the backup file like this:

    nc -l -p 1024 > backup.tar.bz2

  • Then you pipe the tar command without the f flag through netcat on the sending end like this:

    tar -cvj <all those other options> / | nc -q 0 <receiving host> 1024

    • In the above commands 1024 is just a random port number, anything from 1024 and up should work.
  • If all goes well the backup will be piped through the network without touching the file system being read.

• You can also use SSH:

  tar zcvf – /home | ssh <backuphost> “( cat > home_bkp.tar.gz )”

  • In this example:

    • The directory to backup is /home

    • The backup destination is home_bkp.tar.gz on the machine called <backuphost>.

    • The hyphen before /home tells tar to send output to stdout rather than to a file.

    • Adding the ‘p’ option to tar would preserve file permissions.

Restoring

For the purpose of this tutorial we will assume your backup file is stored in the file system root. We will also assume that you have already gained superuser access through sudo.

• Restore your backup:

  tar -xvpzf /backup.tgz -C /

• If you used bz2:

  tar -xvpjf backup.tar.bz2 -C /

  • A brief explanation:

    • The x option tells tar to extract the file.

    • The -C <directory> option tells tar to change to a specific directory before extracting. ” / ” in this example.

    • The p option preserves all file permissions. This is default action for tar when used by the superuser.

  This will overwrite every single file on your partition with the one in the archive.

• The restoration process may take awhile, depending on the size of the archive and the speed of your computer.

• Once the extraction is complete, re-create the directories which were excluded.

  mkdir /proc /lost+found /mnt /sys

• Reboot and everything should be restored to the state of your system when you made the backup.

  It may not be exactly the way it was when you made the backup, because files created after the backing up won’t be deleted.

Restoring over a network

• If you used nc to backup to another computer the commands to restore are:

  • On the sender side, the side that has the backup file:

    cat backup.tar.bz2 | nc -q 0 <receiving host> 1024

  • On the receiving side:

    • Mount the disk (if you are running from a LiveCD)and type:

      nc -l -p 1024 | tar -xvpjf – -C /mnt/disk

      • The – character will tell tar to accept the input from stdin, the pipe.
      • The backup file will be expanded without being saved on the disk on the receiver, like when the backup was made.

      • “-xvpjf is for a .bz2 file, change j to z if you used a tar.gz backup.

Reformatted Partitions

• If you had to format partitions, update the /etc/fstab file after restoring the backup.

  1. Mount the reformatted partitions on a LiveCD.

  2. Open a terminal and type

     blkid

  3. Making note of the UUIDs, edit the /etc/fstab in the restored root partition.

     sudo nano /mnt/disk/etc/fstab

     Change the UUIDs to match the results of your blkid command.

Restoring GRUB

• In most cases restoring GRUB should not be necessary.

  • If you need to reinstall grub GRUB see this page.

Additional resources

• “Backing Up Ubuntu”

• “Wei’s world: HOWTO: backup my Ubuntu”

1. You’ll need to connect the new drive onto the IDE cable. Most IDE cables have two plugs for hard drives. The headache comes with determining which drive is a master and which a slave, so the computer knows which one to boot to. No two hard drive manufacturers have the same standard for setting those tiny little jumper pins on the unit, so you’ll have to read whatever diagram they have and fiddle with tweezers moving them around and such.

Also, most PCs allow you to select boot order from the BIOS, or with what order you go by on the cables. Our troubles begin when we have a drive with the pins in the wrong place and a BIOS that disagrees with it. You’ll know that you have it all straightened out when you have your original drive as “/dev/hda” and your new drive as “/dev/hdb”. Start it up and test for this, then shut it back down. (Lyle’s edit: Often these days, hard disks are now labelled /dev/sdX, where X are letters of the alphabet, starting with ‘a’.)

2. Disconnect the original drive (so you don’t accidentally mess it up!) and boot the machine from your choice of any live CD distro or “Tom’s root-boot floppy“. Tomsrtbt has the benefit of being fast, having all the tools you need for disk maintenance, and virtually guaranteed to run on any old toaster you have. It’s saved my bacon dozens of times, so another big, whooping shout out to Tom. (Lyle’s edit: You can also use GParted’s Live-CD, very useful).

3. If it’s a used drive, use “fdisk /dev/hdb” to delete partitions. fdisk is really simple. The commands are all one-letter.

• m – The “manual”, it will print all the commands available.
• p – Prints the partition table.
• d – Deletes a partition. It will prompt you for a number, if there’s more than one.
• w – Will write the changes to the drive permanently.
• q – Quits fdisk.

After this, I always use “dd if=/dev/zero of=/dev/hdb” to wipe the new hard drive completely. If it’s factory-new, you can skip these steps. Using “dd” will take a while for a large drive. Figure roughly one hour per ten Gigabytes. Leave it run while you go play Sudoku.

4. Now you will want to set up the new drive’s partitions with fdisk. Once again, use “fdisk /dev/hdb” and type your letters; Linux distros will usually need partition #1 to be the boot section, #2 to be swap, and #3 to be the “/” root directory. But these differ. Some distos want to split up the root partition into multiple sub-partitions, with “/usr/” or “/home/” going on a separate partition.

• n – New partition. You will be prompted to answer whether it’s to be a primary or extended partition, enter a partition number, and specify the beginning and ending blocks. For anything less than four partitions per disk, make them all primary.
• a – make a partition “bootAble”. Usually only the first partition – boot, naturally!
• t – Change a partition’s system Type. By default, fdisk on Linux will make every partition type #83 – native Linux. The only one you will have to change is the swap one, to #82.

Here is a typical Linux setup from a different machine:

Disk /dev/hda: 80.0 GB, 80026361856 bytes
255 heads, 63 sectors/track, 9729 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

   Device Boot      Start         End      Blocks   Id  System
/dev/hda1   *           1           9       72261   83  Linux
/dev/hda2              10         140     1052257+  82  Linux swap
/dev/hda3             141        1550    11325825   83  Linux
/dev/hda4            1551        9729    65697817+  83  Linux

Check the start and end cylinders for each partition. The boot partition (#1) only needs to go from 1 to about 9, since the boot section only needs a few MBs. The swap partition… is the subject of controversy. Nobody ever knows exactly what to make it. Just take a guess close to what most systems seem to install by default and go from there. The rule of thumb is “twice your RAM size”, but in real life I’ve found that the minimum you need should be about 256MB (1/4th of a Gig) and the most about a Gig. For a home system, anyway.

Because I have to have huge amounts of disk space for home (backing up the website, archiving my graphics art and writing work, emulating systems), I have 62 Gigabytes here for /home/ and only 11 Gigs for the “/” partition, which is extremely generous for most Linux installs even with more stuff on it than anybody would need.

5. Anyway. Now that you have the disk partitioned, make a file system on all partitions with type #83. “mke2fs /dev/hda1″ and do the same for /dev/hda3… etc.

6. Now you’re ready to shut down, reconnect the main hard drive, and boot back to your main drive. Mount each partition from both disks in tandem – “mount -t ext2 /dev/hda1 /mnt/hda1″ and “mount -t ext2 /dev/hdb1 /mnt/hdb1″ for instance. Then copy everything from one disk to the other, one set of partitions at a time, repeating for hda3, hda4, etc. If you use “cp -pR /mnt/hda1 /mnt/hdb1″, that will copy everything Recursively (R) (all folders, subfolders, and files for the whole file tree), and preserving permissions (p). Do not copy the swap partition, of course!

7. And… one last step. You need to use either Grub or Lilo to write to the Master Boot Record of the new hard drive. I use Grub, since… well, I dunno, I like it! And it’s flexible. The only problem with grub is that its syntax is so alien. Here’s a good tutorial on Grub. My only habit is to mount the old boot partition and make a copy of the menu.lst file to somewhere else, then after I use Grub to create the bootable record on the new disk, I copy over the menu.lst file to it. Providing, that is, that the old system also used Grub and I’m not changing anything.

8. Finally, shut down yet again, remove the old hard drive, change the new hard drive so it will be the master disk instead of the slave, checking jumper pin arrangement on the drive, cable position on the IDE cable, and BIOS boot order. Congratulations, you did it! But do check to make sure the system is stable and no file data has been lost before you go wiping the old hard drive… this is why we copied files instead of moving them!

This was just a quick overview of the process. There’s a lot of details for each of the steps. Consult documentation for specifics, and if you’re not at least competent with the tools involved, try practicing on a scratch monkey computer until you get the hang of it without risking your valuable data.

And be sure to check the comments for all of the corrections left by visitors pointing out where I goofed this up!

1) To rename filenames with CAPS characters to lowercase:

rename ‘y/A-Z/a-z/’ *

2) To remove <spaces> from filenames:

rename ’s/ /-/g’ *

3) To strip files with the .bak extension:

rename ’s/\.bak$//’ *.bak

Free: The free command displays the amount of free and used memory in the system. For example, free -m gives the information using megabytes, which is probably most useful for current computers.

lsb_release -a: The lsb_release command with the -a option prints version information for the Linux release you’re running

sed: The sed (or Stream EDitor) command allows search and replace of a particular string in a file. For example, if you want to find the string “cat” and replace it with “dog” in a file named pets, type
sed s/cat/dog/g pets.

grep: An excellent tutorial.

Running commands sequentially in one command:
If you need to execute multiple commands in sequence but don’t need to pass output between them, there are two options based on whether or not you want the subsequent commands to run only if the previous commands succeed or not. If you want the commands to run one after the other regardless of whether or not preceding commands succeed, place a ; between the commands.

For example, if you want to get information about your hardware, you could run lspci ; lsusb which would output information on your PCI buses and USB devices in sequence.

However, if you need to conditionally run the commands based on whether the previous command has succeeded, insert && between commands. An example of this is building a program from source, which is traditionally done with ./configure, make, and make install.

The commands make and make install require that the previous commands have completed successfully, so you would use ./configure && make && sudo make install. Usually you’ll need an SUDO, for the last command, so make sure you’ve authenticated SUDO prior to running that command.

Learning the Linux Shell: A great tutorial.

last edited 2008-07-16 22:14:39 by JohnCrawford

Tips from IBM, not all that complicated, but worth looking at.

Other tips include…

• Sharing computers on a Linux (or heterogeneous) network.
• IBM’s top 10 Linux articles and tutorials.
• Miscellaneous tips.
• Linux-powered networking: Part 1, Part 2 and Part 3.
• Industrial-strength Linux box lockdown, a howto: Part 1 and Part 2.
• Introduction to cryptology concepts: Part 1, Part 2 and Part 3.

Note: Some of these links requires registration.

Essentially RTorrent, CTorrent and Transmission-CLI.

This is an excellent treatment on advanced file permissions in Linux, though centered on OpenSUSE, it’s still very helpful and applicable to any distro.

Though John credits this source for working with advanced Linux permissions in Ubuntu under Nautilus (Gnome).

2. The watch command.

3. Command line intro summary, with some rarely used switches (TLDP).

4. If you’re using a system which has a lot of users, and you’d like to see who has started a particular script, daemon, or binary, then the pstree utility is very helpful. It draws a tree of all currently running processes – allowing you to see which processes are related.

SSHMenu is much like Putty, in that you can set up individual address book style destinations to SSH into, and save settings for each. While scripting them from the CLI and applying a chmod +x, I can see the benefit of using a centralized GUI.

From the site:

Each menu option will open an SSH session in a new terminal window. You can organise groups of hosts with separator bars or sub-menus. You can even open all the connections on a submenu (in separate windows or tabs) with one click.

Here’s a killer feature: imagine if every time you connected to a production server the terminal window had a red-tinted background, to remind you to tread carefully. Using terminal profiles, SSHMenu allows you to specify colours, fonts, transparency and a variety of other settings on a per-connection basis. You can even set window size and position.

It also leverages Gnome terminal profiles, so you can set up unique profiles to specify transparency, text color and background to remind you of which SSH Host you’re in. It’s written in Ruby, so once you add the source to your /etc/apt/sources.list you’ll need to sudo apt-get update, and import their GPG keys, and re-update again, then install.

Sources: Linux.com, SSHMenu.

HOW TO’s:

Ubuntu Studio 8.04 Desktop:

Ubuntu Studio 8.04 Server:

–

Ubuntu Studio homepage:

I expect the concept of the on-chip OS to catch on quickly, and if they ultimately offer package management (such as APT) and allow a /home partition on a hard disk (or even a solid state drive!) this could become extremely popular.

ASUS has branded Splashtop “Express Gate”. It will run off a chip, and will be an instant-on OS. At any time you can click a reset button on screen and boot up to a “traditional” OS off a hard disk. It even recognizes wireless interfaces!

If they’d allow for a VNC/RDP client — then this would be a an amazing offering as a base OS on a mini laptop to compete with the likes of the EEE PC.

Source: Splashtop (manufacturer blog article).

Technically speaking, ReiserFS is an amazing file system, my preferred file system for all my external drives and internal systems. While Reiser4 was being worked on, it’s not ready for primetime and likely never will be ready. It appears other large distros like SuSE have switched back to ext3 since ReiserFS may become a dead filesystem now.

I feel that ReiserFS is itself a tried and tested file system, though I do have great concern about this development. I never cared for the performance of ext3 and while ext4 is in development, it will be some time before it’s ready.

For my own systems I plan to stick with ReiserFS for now (not reiser4) and perhaps consider XFS or JFS (my research shows XFS to be one of the better file systems, but I have not yet personally tested it.)

Here are a few good artciles on the possible fate of ReiserFS due to Hans Reiser’s current real life issues.

1. ZD Blogs

2. IT News.au

3. Daily Tech

4. CNN

Here’s his initial post:

Linus Torvalds on 2008-03-31 15:37 EST

Description of problem:

youtube no workee – fedora 9 not usable for wife

Version-Release number of selected component (if applicable):

swfdec.x86_64 0.6.2-1.fc9
swfdec-gtk.x86_64 0.6.2-1.fc9
swfdec-mozilla.x86_64 0.6.0-1.fc9

How reproducible:

I didn’t try a lot of videos, but I couldn’t find a single one that actually
worked. And what’s the internet without the rick-roll?

Some just show a light gray background, some give the play buttons etc, but show only a black screen even when the red ball at the bottom moves along..

Steps to Reproduce:
1. Install current Fedora 9
2. Rick-roll!
3. No profit!

Actual results:

Some videos just show a light gray background, some give the play buttons etc, but in the latter case show only a black screen even when the red ball at the bottom seems to moves along..

Expected results:

Rick Astley in all his glory! People have reported that youtube videos are supposed to work with swfdec, so I presume they have worked at some point and have been broken recently.

Just to test that this isn’t just a anti-rick-roll security feature, I also tested some other videos, but let’s face it – we do need Rick for the “Full Internet Experience”.

Additional info:

This is “high” priority because the wife will kill me if she doesn’t have her videos. And the adobe player won’t install on current rawhide due to some library issues.

“Obi-wan Kenobi, you’re our only hope”

I can imagine his wife standing there, arms crossed, “You wrote the Linux Kernel, and you can’t get my YouTube videos to play!!! (Actually he only wrote about 2% of what is now the current Linux Kernel, but that wouldn’t make it any funnier).

The rest of the forum thread can be read here.

Essentially, it strips away the need to run a whole RDP desktop (via Rdesktop in Linux). Instead, you can execute an automated login to a windows box via RDP, then the will quickly collapse, running the single-application you actually need to run, with just that window showing up on your Gnome or KDE desktop; similar to what citrix does when stripping applications to have a “seamless” overlay on a desktop. This would be scripted in a shell command . . .

rdesktop -A -s "c:\seamlessrdp\seamlessrdpshell.exe notepad" -u username -p password

(You could also leave the password blank, but then you’d have to type it manually).

Click here for SeamlessRDP.

Here is also an interesting article in Linux.com explaining how one could strip the whole desktop leaving just the windows start menu to draw on your Linux desktop, rather interesting possibilities when you absolutely must play with Bill’s OS.

One important tip: Make sure the user you’re using to log into the windows box has its RDP default session set to END the session after a disconnect! SeamlessRDP can sometimes end an RDP session abruptly when closing an application, leaving the session open. If the specific user used to login to the RDP session is not set to auto-end the session, you could lock yourself out of a subsequent session.

EDIT: As of March 26, 2008 the Linux Reality podcast had its 100th and final episode. The series is really great and any user new to Linux will enjoy nearly every episode in the series. The IRC Channel (#linuxreality on irc.freenode.net) and the forums are still quite active. Cheers Chess, you were great and it’s great to chat with you on the IRC!

A great podcast show for those interested in Linux topics. The production quality is good and Chess Griffin is a talented host. Many of the topics are geared towards half-newbies, though some are more in depth. There’s a very large library of prior podcasts discussing a large variety of topics and some very interesting interviews, so there’s plenty to load up on your mp3 or ogg player of choice to enjoy.

From the article:

Thompson supervises a payment system that operates on about 4,000 servers running Red Hat Linux in the same manner that eBay and Google conduct their business on top of a grid of Linux servers. “I have been pleasantly surprised at how much we’ve been able to do with this approach. It operates like a mainframe,” he said.

…

PayPal takes Red Hat Enterprise Linux and strips out all features unnecessary to its business, then adds proprietary extensions around security. Another virtue of the grid is that PayPal’s 800 engineers can all get a copy of that customized system on their development desktops, run tests on their raw software as they work, and develop to PayPal’s needs faster because they’re working in the target environment. That’s harder to do when the core of the data center consists of large Unix symmetrical multiprocessing boxes or mainframes. In neither case is it cheap to install duplicates for developers, he said.

PayPal “pays very close attention to the Linux kernel development process” lead by Linus Torvalds and the kernel maintainers because future capabilities are being debated and resolved through the process, he said.

PayPal has experimented with virtualization and is watching carefully developments in open source virtualization, still a young field. “One place we see the kernel process at work is in virtualization,” Thompson said. VMware’s ESX Server can run Linux, as can the open source Xen hypervisor; both work outside the Linux kernel but can be linked to its internal operations. A year ago, Torvalds approved the addition of a contributed Kernel Virtual Machine, which runs inside the kernel and makes use of the kernel’s own memory management and other functions.

“If we could fully virtualize our middle tier, that would be another step of cost advantage,” said Thompson. More fully virtualized data centers also would allow him to shift workloads across the grid, depending on time of day and traffic volumes, which would lead to additional savings.

Source: Information Week.