Matt's Gnu-linux help, tutorials and advice.

Flash broken in firefox 4 in Arch Linux..... for some 64bit users

2011-04-09T23:00:00.001+01:00

Firstly my apologies for not posting in such a long time. I am now at Uni and have been quite busy. Also I haven't really been doing much in the way of Linux fixes recently as things have been........ well just working!

Until recently that is. I saw the mockups of firefox 4 and liked it instantly. The second it was in the Arch Linux repositories I installed it and started using it as my main browser instead of opera or chromium.
Sadly there are some issues with it, especially with flash. The other issues were just annoyances that were fixed with about:config changes. (Not that I'm blaming the folks at Mozilla for this, there is a high chance its Adobe's fault...)

For some reason some websites using flash display fine but others go haywire and parts of the flash disappear or flicker.

One suggested solution I found was to do the following:

1. Click Here
2. right click on player
3. click settings
4. choose left-most tab
5. uncheck "Enable Hardware Acceleration"

However this only worked for some people but not for me.

There is a pre-release copy of Adobe Flash Square in the archlinuxfr repository (I presume most Arch Linux users have this for yaourt), and installing this, which prompted me to remove the existing copy of flash, and restarting firefox fixed the issue for me completely.

For those with archlinuxfr:

bash$ sudo pacman -S flashplugin-prerelease

or to get archlinuxfr add the following to your /etc/pacman.conf in the appropriate section.

[archlinuxfr]
# The french Arch Linux communities packages.
Server = http://repo.archlinux.fr/$arch

EDIT: After a security hole was discovered the package was pulled but due to a new release, the package is back in the AUR as flashplugin-prerelease at version 11.0.1.60-2 as of today. I'm unsure as to what version is in the archlinuxfr repository currently but I assume it will be older than the one in the AUR.

Fix popping sounds in Linux

2010-11-11T20:05:00.003Z

Quite a lot of people have had a problem when, when their computer boots up and just before sound plays/stops they get loud popping noises which can be very annoying and a bit concerning.

This is caused by a new 'feature'which is supposed to save power by powering down the sound chip when not in use.

To prevent it from doing this thus preventing the popping sounds edit your:

/etc/modprobe.d/modprobe.conf

(I prefer to use vim. Nano is a good alternative)

bash$ sudo vim /etc/modprobe.d/modprobe.conf

and add the line:

options snd-hda-intel power_save=0

then reboot, and your issue should be fixed.

This only works for intel chips using the snd-hda-intel driver.

To look for options for other sounds cards with the same issue ( I haven't heard of any) open a terminal and type:

modinfo snd-MY-MODULE

replacing MY-MODULE with the sound module used.

Saitek cyborg R.A.T in Linux

2010-10-30T12:44:00.000+01:00

I have finally finished setting up my new system with arch linux.

I am currently having problems with sound not muting when headphones are plugged in so this may well be one of the topics I cover shortly.

I did however manage to get my new mouse, the saitek cyborg R.A.T.5 running in Linux.

There is a problem when you first plug it in that means you have to click the change mode button every second click but I managed to find a fix that should also work for the R.A.T.3 and the R.A.T.7.

Here are pictures of the R.A.T.3 ,5 & 7 for those who haven't seen them:

The R.A.T.3:

The R.A.T.5:

The R.A.T.7:

(there is also a R.A.T.9 but I can't confirm that the fix works for it)

The fix is quite simple.

It does rely on Xmodmap.

Here are instructions using sudo. If you have a root account ommit sudo and run them as root.

First you need to edit the Xmodmap file with your favourite editor. I prefer vim. the Xmodmap file should be here:

/etc/X11/Xmodmap

so to edit with vim I typed:

bash$ sudo vim /etc/X11/Xmodmap

If you are unused to vim I recommend nano as a good alternative:

bash$ sudo nano /etc/X11/Xmodmap

All the file has to contain is:

pointer = 1 2 3 4 5 6 7 8 9 10 11 12 0 0 0

If there is already information in your Xmodmap I recommend putting this at the start.

You can then restart X or your computer and it should work without a problem.

Announcement: Ubuntu 10.10 Released

2010-10-13T13:29:00.005+01:00

Ubuntu 10.10 was released on Sunday and is being hailed as the best Ubuntu yet.

It seems to be even faster than 10.04 and is much more stable.

A lot of hardware that has been problematic in older versions of Ubuntu has sprung back to life and the release seems generally good.

Have a look at the features HERE.

Download a copy HERE

If you have Ubuntu 10.04 installed you can directly upgrade to 10.10.

Start System/Administration/Update Manager.
Click the Check button to check for new updates.
If there are any updates to install, use the Install Updates button to install them, and press Check again after that is complete.
A message will appear informing you of the availability of the new release.
Click Upgrade.
Follow the on-screen instructions.

Of course on the forums for a few weeks there will be posts claiming that it is full of bugs and more useless than ever etc....

Every release has some bugs when its brought out but they usually get fixed very quickly.

De/Wm's: GNOME

2010-10-07T09:00:00.006+01:00

GNOME stands for "GNU Network Object Model Environment" and is a desktop environment for various Unix-like operating systems, most notably GNU/Linux, and as part of the Java Desktop System in Solaris.

Gnome is very popular, rivalled by KDE, and is used as the Default Desktop Enviroment on many LInux Distro's.
Gnome is written in C mainly as this is the language GTK+ is written in.

The Gnome desktop enviroment usually comes with the following software:

X window manager--->Metacity
X display manager---->GDM
File manager------------>Nautilus
Widget toolkit --------->GTK+
Terminal emulator ->GNOME Terminal
Text editor-------------->gedit
Video player----------->Totem
Audio player----------->Rhythmbox
CD burners------------->Brasero
Image viewer---------->Eye of GNOME
Office suite-------------->GNOME Office
Web browser----------->Epiphany
E-mail client------------->Evolution
Instant messenger---->Empathy
Archive manager------>File Roller
PDF viewer--------------->Evince
IDE---------------------------->Anjuta
Widget engine----------->gDesklets
Licenses used------------> GPL, LGPL

Developers of OS's often modify the software bundled with the desktop enviroment in the repositories and often the desktop enviroment itself to include branding, for instance Ubuntu.

Here is a screenshot of gnome 2.3 running.
(Click to see full size image)

Announcement: Desktop Enviroments & Window Managers

2010-10-06T19:31:00.003+01:00

Most of the content in this blog so far has mainly been shell script or Non GUI tutorials.

I have decided to start doing GUI tutorials as well.

I will be starting with information about Desktop environments/Window managers in Linux (De/Wm for short) by having a brief blog post on each, prefixed with: "De/Wm's:".

I may not cover all De/Wm's but will certainly try to cover quite a few, hopefully including:

DE's:
Gnome
KDE
XFCE
LXDE

WM's:
IceWM
Openbox
Fluxbox
Beryl
Compiz
Metacity
KWin

There are of course many more.

A window manager is system software that controls the placement and appearance of windows within a windowing system in a graphical user interface.
A desktop environment is a collection of software designed to give functionality and a certain look and feel to an operating system.

A Desktop Environment will have a window manager, however it it possible to run a window manager without a Desktop Environment..... up to a point.

Here are the default Wm's for some De's:

---DE-------WM---
Gnome---->Metacity
KDE------>KWin
XFCE----->XFwm
LXDE----->Openbox

Different Desktop environments and window managers are written in different programming languages and are designed for different purposes.

I will cover this in upcoming articles.

64 Bit flash returns to linux!

2010-10-05T15:39:00.000+01:00

Sorry not to have posted for so long. I have been very busy.

Anyway, I am a 64 bit user and have had no issues with anything apart from flash.

I have tried the 32 bit plugin with ndiswrapper which was awful and very slow to respond. I have also tried several open source alternatives such as Gnash and although they are coming along well they don't seem to be up to much as of yet.

A while ago Adobe released a 64 bit plugin but it was scrapped and removed from their site due to an apparent security hole.

However it is now back and apparently fixed so I have tried it and it works reasonably well.

It has been nicknamed "Square" and is availible for Windows, Mac and Linux.

The only noticable difference is that flash videos are no longer stored in /tmp/ while streaming but are now in your browser's cache, for firefox users this would be /home/user/.mozilla/firefox/profile.default/Cache/ .

For Ubuntu users there is a PPA for it. To install use:

bash$ sudo add-apt-repository ppa:sevenmachines/flash

bash$ sudo apt-get update && sudo apt-get install flashplugin64-installer

For other distro's or for those who want it directly from adobe labs the download link is:

labs.adobe.com/technologies/flashplayer10/

Installation of the Flash preview has to be done manually – but this isn’t that difficult.

Unpackage the file.
Copy the libflashplayer.so file to the plugins folder of your browser:
/usr/lib/mozilla/plugins/

I hope that works for 64 bit users out there. Its certainly much smoother for me.

Users should note that as this release is termed a ‘preview’ and should not be taken as a final stable release so may be slightly buggy.

Simple command line usage

2010-09-22T18:35:00.002+01:00

I have been using the linux command line for some time and often when I post help I miss some steps as I take them as already known.

Usually most linux users will know them but noobs to linux may struggle and give up.

Here are a few simple commands that you may use quite often and a small explaination of what they do.

ls:
bash$ ls
This simply LiSts the files in the current directory. (If you have just opened the terminal this will be your home folder)
Here are some of the arguments you can use:
-a list hidden files
-d list the name of the current directory
-l long listing giving details about files and directories
-R list all subdirectories encountered
-t sort by time modified instead of name
There are more.

cd:
bash$ cd /Path/To/Directory
This allows you to change what directory you are in. You can cd into any directory you want.
For instance if

bash$ ls
Test

then you can:
bash$ cd Test
to go into directory Test.
You can also use the full location for instance:
bash$ cd /home/user/test
Additionally
bash$ cd ..
will take you to the directoy above the one you are browing.

mv:
This is used to change the name of a directory
Type mv followed by the current name of a directory and the new name of the directory.
bash$ mv TEST NEW-NAME-FOR-TEST

cp:
This is used to copy files.
bash$ cp FILE FILE2

mkdir:
This is the command to make a directory.
For instance:
bash$ mkdir TEST
would make a directory named TEST.

pwd:
This will show you the full path to the directory you are currently in.
This is very handy to use, especially when performing some of the other commands on this page
bash$ pwd

rm:
This is the ReMove command. You can only use this to delete individual files.
bash$ rm FILE
To delete a folder you must add args -r and -f for instance:
bash$ rm -rf TEST
but this permenantly removes files so is dangerous.
To be safer add the -i arg which will ask you if you're sure.

sudo:
In linux you are simply a user, not administrator. The administrator account is called the root account, but using this accound can be very dangerous.
For instance if you log in as root and type
bash$ rm -rf/
it will attempt to wipe the computer.
This is why sudo is used. Sudo lets you run a command as root from a user account.
Once sudo is set up (done automatically on most distros) you can simply:
bash$ sudo COMMAND-TO-RUN-AS-ROOT -ARGS
But be careful as:
bash$ sudo rm -rf /
is just as destructive as when logged in as root.

./
This is used to run a file that is not installed as a program. The file must have permissions to run.
bash$ ./FILE-TO-RUN

chmod
This is used to alter file permissions.
Usage is:
bash$ chmod [+/- x] [0-7 0-7 0-7]
A numeric mode is from one to four octal digits (0-7), derived by adding up the bits with values 4, 2, and 1. Omitted digits are assumed to be leading zeros. The first digit selects the set user ID (4) and set group ID (2) and restricted deletion or sticky (1) attributes. The second digit selects permissions for the user who owns the file: read (4), write (2), and execute (1); the third selects permissions for other users in the file's group, with the same values; and the fourth for other users not in the file's group, with the same values.

THat should get you used to command line. You can use:

bash$ COMMAND --help
or
bash$ help COMMAND
or
bash$ man -k COMMAND
to view the help for a command.

A good list of more commands is HERE

How to set up passwordless ssh

2010-09-19T11:00:00.005+01:00

If you have a server or remote machine that you ssh into a lot or want to run a backup script to, you probably have to type in your ssh password every time.

If its no-body but you that has access to your user account on the local machine you can set up passwordless ssh between the local machine and the server securely.

This is done with rsa keys.

First you need to make a public key on the local machine.

bash$ ssh-keygen -t rsa

use -t rsa unless you plan on accessing an older machine or a machine with an older version of OpenSSH.

Do NOT enter a passphrase.

Now you need to copy the public key to the remote machine's list of authorized keys.

bash$ ssh-copy-id -i ~/.ssh/id_rsa.pub USERNAME@SERVER

Of course, you should use your custom public key if you did not use the default name in the previous step.

After executing this, it will ask you for your password, this is just the ssh password to the remote machine for the username that you used.

Upon completion, there should be a file on the remote machine ~/.ssh/authorized_keys that contains the public key that you just generated.

You can test out if you were successful now by ssh'ing to the remote machine, you should no longer be asked for a password.

However this is potentially insecure as if someone managed to get hold of your public key they could get into your machine.

To prevent this you can edit the file .ssh/authorized_keys and add from="LOCAL-IP-ADDRESS" to the beginning, leaving a space before ssh-rsa and replacing LOCAL-IP-ADDRESS with your machines IP. This will of course only work with static IP's.

You can further restrict what can be done with passwordless ssh by adding any of the following options to the beginning of the authourised_keys file:

no-port-forwarding,no-X11-forwarding,no-agent-forwarding,no-pty

This will still not let you run passwordless ssh from cron. If you wish to do this I recommand THIS tutorial.

(Some content from patsissons on the ubuntu forums)

Secure SHell FileSystem - SSHFS

2010-09-18T17:17:00.004+01:00

SSHFS (SSH FileSystem or Secure SHell FileSystem) is a filesystem client to mount and interact with directories and files located on a remote server or workstation.
The client interacts with the remote file system via SFTP.

SSHFS works on most UNIX and Linux systems. The current implimentation is based on
FUSE.

As it uses ssh it is more secure than samba.

To run sshfs you need to install fuse as well. If you install SSHFS through a package manager it will install all dependancies.

For example:

bash$ sudo pacman -S sshfs (Arch)

bash$ sudo apt-get install sshfs (Debian and derivatives)

Once installed, you can mount any location on a remote server that you have ssh access to.

You need to have made the mountpoint (The location the remote directory will be mounted) first:

bash$ mkdir /path/to/MOUNTPOINT

You may then mount the remote directory into the mountpoint. This is done by:

bash$ sshfs USERNAME@SERVER:LOCATION MOUNTPOINT

The mountpoint now appears to contain the contents of the mounted directory on the remote server.

You can now interact with the contents as if it was on the local machine but you will be using the permissions of the user you mounted it with.

For example:

bash$ mkdir /server1
bash$ sshfs user@192.168.1.2:~/home/user/ /server1

would be used to mount the directory /home/user on the remote server in directory /server1 on the local machine.

To unmount mounted directories simply use:

bash$ fusermount -u MOUNTPOINT

In my example this would be:

bash$ fusermount -u /server1

OpenSSH - Connect to and manage other machines.

2010-09-17T21:13:00.000+01:00

Most people who use LInux or UNIX will usually recognise and use ssh quite commonly to connect to other machines and run updates or programs and generally monitor and control machines.

OpenSSH (the package including ssh) includes more than just ssh.

OpenSSH is a FREE version of the SSH connectivity tools that technical users of the Internet rely on.
Users of telnet, rlogin, and ftp may not realize that their password is transmitted across the Internet unencrypted, but it is.
OpenSSH encrypts all traffic (including passwords) to effectively eliminate eavesdropping, connection hijacking, and other attacks.

Additionally, OpenSSH provides secure tunneling capabilities and several authentication methods, and supports all SSH protocol versions.

The OpenSSH suite replaces rlogin and telnet with the ssh program, rcp with scp, and ftp with sftp.
Also included is sshd (the server side of the package), and the other utilities like ssh-add, ssh-agent, ssh-keysign, ssh-keyscan, ssh-keygen and sftp-server.

I often use ssh to connect to machines and run programs and backups etc...

SFTP is also useful ( a secure version of FTP) for transferring files.

The client side of ssh is installed with many Linux and UNIX OS's as it is now considered as an important program.

To connect to an ssh server you simply open a terminal and type:

bash$ ssh user@server-address

If the username on the server/ remote machine is the same as the current machine you can simply type:

bash$ ssh server-address

for example

bash$ ssh 192.168.1.2

or

bash$ ssh joebloggs@192.168.1.2

You could also use:

bash$ ssh 192.168.1.2 -l joebloggs

By default ssh uses port 22 so you may have to set this as an exception in your firewall.

You can however pipe ssh through another port with -p.

Also if the server IP is named in your hosts file /etc/hosts, you can then use the server's name instead of IP address.

For example:

bash$ ssh joebloggs@joebloggs-computer

The ssh help dialogue:

bash$ help ssh
usage: ssh [-1246AaCfgKkMNnqsTtVvXxY] [-b bind_address] [-c cipher_spec]
[-D [bind_address:]port] [-e escape_char] [-F configfile]
[-i identity_file] [-L [bind_address:]port:host:hostport]
[-l login_name] [-m mac_spec] [-O ctl_cmd] [-o option] [-p port]
[-R [bind_address:]port:host:hostport] [-S ctl_path]
[-w local_tun[:remote_tun]] [user@]hostname [command]

To set your machine up as an ssh server, to allow connections to your computer, you can simply install package sshd and set it to run as a DAEMON or at startup if your OS doesn't do it automatically.

People will have to know your password to connect to you computer so its quite secure.

Just make sure you use a good password.

(Resources at the top of this log are from: http://www.openssh.com/ )

Bash Aliases

2010-09-16T16:22:00.002+01:00

Linux users who commonly use the terminal often get bored of typing horribly long command sequences with arguments for routine jobs.

This is where aliases can come in useful.

(Presuming you use Bash and not sh)

With aliases you can set a command such as Pacman -suy to the name update, so that when you type bash$update in a shell it runs pacman -Suy.

There is more than one way to do this. You can just add the alises to the .bashrc file in your home folder and if you wish to do it this way I recommend you do it at the end.

The other way is to use a seperate file.

If you wish to use a seperate file for your aliases like me, then carry on reading. If you want to add them directly to the .bashrc then skip down a few steps.

Firstly you must create an alises file if one isn't already created on your system.

The usual name for the alias file is .bash_alises the "." at the beginning making it a hidden file.

You can call it what you want.

You must then set that aliases file to run in the .bashrc.

To do this edit the .bashrc file in your home folder and near the end add:

 if [ -f ~/.bash_aliases ]; then
. ~/.bash_aliases
fi

or if you named your aliases file something different swap .bash_alises for the name of your file.

Some Linux operating systems have this pre-added along with an empty aliases file.

You can now add aliases to your system either in the .bashrc or .bash_alises (or your own) file.
Aliases must always use the correct format :

alias new-command='horribly long command -arguments'

You can also make system wide aliases by adding them to your /etc/bashrc

You can also put a hash [#] after an alias and then type a quick description so you remember what its for.

You can view all aliases in use by typing bash$ alias in a terminal.

For example, some useful aliases pulled from the internet:

 
# ls
alias ls='ls --color' # Use colors
alias lsd="ls */ -d" # Display only directories
alias l="ls -l"   # Long format
alias la="ls -la"  # Long format with hidden files

# cd
alias ..="cd .."  # Drop up a dir with ..
alias ...="cd ..; cd .."

#cp/mv
#Prompt for overwrite, verbose
alias cp="cp -iv"
alias mv="mv -iv"

# misc
alias fproc="ps aux | grep "  # Find running processes
alias cls=clear
alias gop='gnome-open'
alias df="df -h"    # Human readable format
alias du="du -h"    # Human readable format
alias mkdir="mkdir -v"   # Verbose directory creation
alias grep="grep --color"  # Highlight search word
alias oo="ooffice"

# chmod shortcuts
alias "x+"="chmod +x"
alias "x-"="chmod -x"
alias "w+"="chmod +w"
alias "w-"="chmod -w"
alias "r+"="chmod +r"
alias "r-"="chmod -r"

alias shutdown='sudo shutdown -h now'

From the arch wiki for pacman:

 
# Pacman alias examples
alias pacupg='sudo pacman -Syu'        # Synchronize with repositories before upgrading packages that are out of date on the local system.
alias pacin='sudo pacman -S'           # Install specific package(s) from the repositories
alias pacins='sudo pacman -Up'         # Install specific package not from the repositories but from a file 
alias pacre='sudo pacman -R'           # Remove the specified package(s), retaining its configuration(s) and required dependencies
alias pacrem='sudo pacman -Rns'        # Remove the specified package(s), its configuration(s) and unneeded dependencies
alias pacrep='pacman -Si'              # Display information about a given package in the repositories
alias pacreps='pacman -Ss'             # Search for package(s) in the repositories
alias pacloc='pacman -Qi'              # Display information about a given package in the local database
alias paclocs='pacman -Qs'             # Search for package(s) in the local database

Here are some popular ones for ubuntu pulled from the forums:

 
alias agi='sudo apt-get install'
alias agr='sudo apt-get remove'
alias agp='sudo apt-get purge'
alias acs='apt-cache search'
alias agu='sudo apt-get update'
alias agg='sudo apt-get upgrade'

Sadly you cannot use sudo with aliases for example sudo [alias]. However you can do this by adding:
alias sudo='sudo '
to your aliases file.

Of course you can use very diverse names and make some just for fun.

I have seen alias icanhas='sudo apt-get install' several times.

News: Ubuntu 10.10

2010-09-14T20:58:00.002+01:00

Ubuntu 10.10 will be here shortly and it looks like its going to be good.

As usual the first few weeks after the release will be plagued by people claiming that its buggy or that it doesn't work, but this happens every release and it can take a month for all the bugs to get straightened out.

You can get one of the countdown banners for your site here: http://www.ubuntu.com/community/countdown

You can try the Beta before the release here: http://www.ubuntu.com/testing/maverick/beta

I apologize for the lack of posts. I have been unwell but hope to carry on posting as normal from tomorrow.

Set up sendmail in arch linux

2010-09-10T12:46:00.002+01:00

I recently had the need to have my backup script send me an email with the backup logfile when it was done.

Here is how I did it (with much help from others)

First you need to install mailx.

The package: mailx-heirloom seems to work better for some people so I used it instead.

bash$ Sudo pacman -S mailx-heirloom

I then set up the config file:

bash$ sudo vim /etc/nail.rc
bash$ sudo vim /etc/mail.rc
(Updated as the config file changed.)
and put the following into it:

 set sendmail="/usr/bin/mailx"
set smtp=smtp.gmail.com:587
set smtp-use-starttls
set ssl-verify=ignore
set ssl-auth=login
set smtp-auth-user=login@host
set smtp-auth-password=PASSWORD

Obviously this is for gmail. You may have to modify this for other mail hosts. I tried a few others but gmail was the only one that gave consistant posative results.

You can then send email with the following:

bash$ echo "message" | mailx -s [subject] -a /path/to/attachement recipient@host

My script uses rsync for backups. Here is what I came up with:

 #!/bin/bash
DATE=`date +%m-%d-%y`
echo -n "What directory do you want to back up? : "
read DIRECTORY
rsync -avzuh Targetmachine:~/$DIRECTORY . > /home/username/logs/backup-log-$DIRECTORY-$DATE.txt &&
echo "Backup of $DIRECTORY Complete, Log attached" | mailx -s "Backup Complete" -a "/home/username/logs/backup-log-$DIRECTORY-$DATE.txt" user@HOST

It is only made to backup one directory at a time but that is all I currently need. I'm sure more experienced users could modify this and intergrate it as a cron job with passwordless ssh which is what I will be doing when I get my 4 1TB HDD's..

Keep your /etc/pacman.d/mirrorlist up to date!

2010-09-08T15:28:00.007+01:00

I usually update my server weekly and usually get about 2-10 updates.

However for the last few weeks I haven't been getting any updates at all.

I ignored this for a while but then it became a bit concerning.

After a bit of thinking I realised exactly what was wrong. The repos I was using were no longer being updated.

When pacman gets an update for itself it often says that /etc/pacman.d/mirrorlist has been installed as /etc/pacman.d/mirrorlist.pacnew,(probably to conserve your current server choice) but me being lazy, I had just left it that way.

I quickly typed:

bash$ sudo rm -rf /etc/pacman.d/mirrorlist
bash$ sudo mv /etc/pacman.d/mirrorlist.pacnew /etc/pacman.d/mirrorlist

then selected the repos I got the best ping from, and uncommented them.

Ran

bash$ sudo pacman -Suy

Got about 35 updates and felt a bit foolish. At least thats one mistake I won't make again.

Change TTY text and background colour.

2010-09-07T22:57:00.006+01:00

Change TTY text and background colour.

I recently found a way to change the colour of the text in the TTY screens which is kinda neat if you use them a lot. I use TTY's all the time to manage my servers.

The command is setterm.

The two variables for colour change are -foreground and -background.

Here are the possibilities from the help dialogue:

bash$ [ -foreground black|blue|green|cyan|red|magenta|yellow|white|default ]
bash$ [ -background black|blue|green|cyan|red|magenta|yellow|white|default ]

The usage for this command is as follows:

bash$ setterm -background [colour] -foreground [colour]

For instance for a green on black theme I would use:

bash$ setterm -background black -foreground green

and if the background was already black you could just use:

bash$ setterm -foreground green

If you want this to happen every time you log in you could add it to your .bashrc otherwise you need to manually do this every time you log in to a TTY.

Kill unresponsive or misbehaving processes.

2010-09-03T01:21:00.004+01:00

Anyone using linux should know how to do this as it comes in useful more often than you can imagine.

There are multiple ways to do it.

I will post the most common CLI methods.

Firstly the kill command.

You can use kill to terminate a process that is being run by you, or if logged in as root or using sudo you can kill a program run by anybody.

kill has many uses but I will only post the most simple one.

To kill a process using kill you must first have its PID (process ID).

To get this you can use the command pidof.

You may then kill the process.

For instance;

if

bash$ pidof [process]

gives [output]

then you can

bash$ kill [output]

which will end process [process]

For instance:

bash$ pidof rhythmbox
2531
bash$ kill 2531

would be used to kill rhythmbox

This is a rather long method however.

This is where the comand killall comes in.

You do not need to use the PID when using killall.
You can simply use:

bash$ killall [process]

and the process will be killed.

For instance:

bash$ killall rhythmbox

There is another option which lets you see if a program is misbehaving and kill it called Top.

When in a shell simply type:

bash$ top

You get a CLI task manager which shows PID's and cpu usage etc.

Example:

bash$ top

gives me this:

You can kill a process by pressing 'k' then typing in the PID listed for it.

eg:

PID to kill: 2531

it will ask you:

Kill PID 2531 with signal [15]:

and you can simply press enter to kill it or ^C (ctrl+C) to exit the program.

Installing packages with pacman, the safe way.

2010-09-02T17:10:00.004+01:00

On the Arch wiki there is this note:

Note: Do not refresh the package list when installing packages (i.e. pacman -Sy package_name); this can lead to dependency issues. Upgrade explicitly first; before installing new packages.

I will try to explain this further.

This means that the safest way to install a package is to:

bash$  pacman -Syu

bash$ pacman -S [pkg]

As it fully updates your pacman database, fully updates your system then installs your new package from the up to date system.

Often in the wiki the line given is simply;

bash$ pacman -S [pkg]

As this saves time and space and is usually guaranteed to work.

However this is not as good as the first option as it only installs the newest version of the package in your local database which may be out of date itself. It is safe however and it doesn't waste time syncing so is faster.

Some users however have been advising people to use:

bash$ pacman -Sy [pkg]

and even changing wiki articles to do this.

However this is potentially very dangerous as it updates the local database then installs [pkg] from it. The new version [pkg] may pull down updated dependencies that don't work with the packages currently installed thus breaking them. Alternatively it could bring down no new dependencies but depend on files that are newer than the ones installed. (It could of course all go smoothly but I wouldn't want you to risk it)

More experienced users may be happy to live with this risk and fix any dependencies that are broken themselves. However doing this could cause a lot of problems for less experienced users.

I will always advise people to use:

bash$ pacman -S [pkg]

and leave it up to them whether they want to run;

bash$ pacman -Syu

first or not.

Happily in pacman-3.4 and newer you are able to do:

bash$ pacman -Syu [pkg]

which updates your database, system, then installs your program one line.

[If you break packages you can always just use:
bash$ pacman -Syu
but you may have to boot from a live cd to do this if it is more important libs you have broken.

Ubuntu hangs with initramfs error after kernel update.

2010-09-02T16:23:00.005+01:00

Recently I have seen several posts where users have updated their ubuntu systems but when they reboot the system drops to a prompt with somthing similar to the following:

 udevadm trigger is not permitted while udev is unconfigured
udevadm settle is not permitted while udev is unconfigured
udevadm settle is not permitted while udev is unconfigured
udevadm settle is not permitted while udev is unconfigured

ALERT! /dev/disk/by-uiid/939af864-c1a8-41d7-9b24-91d25685b6 does not exist. Dropping to shell

Busybox v1.13.3 (Ubuntu 1:1.13:3-1ubuntu11 built-in shell (ash).

Enter 'help' for built-in commands

initramfs

Several people have blamed GRUB and messed up their systems trying to fix it when the real fix is quite simple.

To fix it you must first understand the problem.

If the apt-get upgrade or the update manger does not successfully complete then a flag is set and update-initramfs does not get to run, hence grub does not have a complete entry on how to boot the new kernel.

To fix it you must first boot into linux. You can do this by booting your last working kernel in GRUB.

To get to grub in ubuntu 9.10 or above hold the SHIFT key as you boot or in versions 9.04 and earlier press Esc when asked to in the top left corner of the screen.

You should be able to boot using the old kernel, probably the second or third entry down. (NOT the recovery menu)

Once you log in you can then open a terminal and type:

bash$ sudo update-initramfs –u -k kernel-release

(replacing "kernel-release" with the newest kernel that wouldn't boot)

If you don't know the kernel that fails to boot then you can use:

bash$ sudo update-initramfs –u -k all

instead but if it fails to release this command you could break all your kernels although this is unlikely.

Once you have done either of the above you can then run

bash$ sudo dpkg–reconfigure

That should clear the upgrade flags and fix any broken packages.

I recommend you then run this (although not necessary):

bash$ sudo apt-get autoclean ; sudo apt-get check

That should cleanup your system and fix any broken dependencies and packages.

Convert flv to mp4 video via ffmpeg

2010-09-02T00:23:00.002+01:00

Quite often people download flash videos in FLV format from youtube or another site and then want to convert them to mp4 to put on an mp4 player or ipod.

It is quite simple to do in a GUI with a program such as k9copy but is a bit more difficult in the command line. Most other command line tutorials for this purpose need a custom version of ffmpeg compiled from source which can get messy.

If you have ffmpeg and all the common codecs installed then use the following:

bash$  ffmpeg -i input.FLV -acodec libfaac -ab 96k -ac 2 -vcodec libx264 -vpre hq -vpre ipod320 -threads 0 -crf 22 output.mp4

obviously replacing input.flv and output.mp4 with your filenames.

This is not very refined and I'm sure with some work more experienced linux users could change this to get better quality. however it does work for most.

If you don't have ffmpeg or any codecs installed then install them. Instructions are available in the wiki for your distro.

Convert Flv to mp3 using FFmpeg

2010-09-02T00:11:00.002+01:00

I have seen a lot of threads on various forums where people ask how they can convert flash videos, downloaded from the Internet, to mp3.

It is quite simple to do and there are many GUI tools for it. However if you want to do it in a script, or simply prefer to use the command line like me, then to following works quite well.

If you have ffmpeg and all the common codecs installed then use the following:

bash$ ffmpeg -i input.flv -ac 2 -ar 44100 -ab 320 output.mp3

obviously replacing input.flv and output.mp3 with your filenames.

If you don't have ffmpeg or any codecs installed then install them. Instructions are availible in the wiki for your distro.

I recommend you read your distro's wiki before installing codecs as there are legal issues in some countries.

Ubuntu = (You may need to enable the medibuntu repository: https://help.ubuntu.com/community/Medibuntu )

then use: bash$ sudo apt-get install ubuntu-restricted-extras ffmpeg

Arch = bash$ pacman -S gstreamer0.10-{base,good,bad,ugly}-plugins gstreamer0.10-ffmpeg ffmpeg

Stress testing in linux

2010-09-01T17:07:00.003+01:00

Today I recieved some experimental thermal compound for testing. To say that it performed amazingly would be an understatement. However I'm not going to name it until it goes on sale.

I decided that to properly test it I would have to run stress tests and properly measure the temps.

Note: Although you can stress test a newly overclocked system under Linux, it is strongly recommended that you don't do it. Why? You can experience data loss from the HDD depending on how your system fails in the event of an unstable overclock. Linux is less forgiving that Windows for rebooting from a hardlocked system without damage to the file system. If you insist on testing various overclocked levels under Linux, it is advisable to backup ALL important data first. Better yet is to install Arch to an old HDD and unplug your HDD(s) that carry your important data to iron out the overclock settings and arrive at a stable system.

To measure the temps I installed HddTemp and Lm-Sensors. I will give instructions for doing this in Arch Linux but I'm sure these packages are available for ubuntu and other Linux Distros in their respective repos.

To install hddTemp and Lm_sensors:

bash$ sudo pacman -S lm_sensors hddtemp

(HddTemp is not really necessary but its always a good idea to keep an eye on your hard drive temperature.) Read HERE for information on use.

You then need to set up Lm_Sensors:

bash$ sudo sensors-detect

And answer yes to all the questions.

Next you need to start the service:

bash$sudo /etc/rc.d/sensors start

I recommend that you automatically load the kernel modules at boot time by adding sensors to the DAEMONS array in /etc/rc.conf

(I Believe this is done automatically in ubuntu. )

eg: DAEMONS=(syslog-ng crond ... sensors ...)

You can then simply type:

bash$ sensors

to show your system temperatures.

Ignore the ALARM's etc as the program needs refining for some chipsets.

Now to set up a stress testing program. Prime 95 is known to be the best for windows and there is a version created for linux.

Firstly open your editor and paste the following into it:

#!/bin/bash

echo "Downloading MPrime . Please wait. This may take some time........" ;

echo ;

wget ftp://mersenne.org/gimps/mprime2511.tar.gz ;

echo ;

 echo "Installing mprime ............." ;
tar  -xvzf

mprime2511.tar.gz ;
rm -rf mprime2511.tar.gz ;
sudo  cp mprime /usr/bin/ ;
sudo chmod 755 /usr/bin/mprime ;

echo ;

 echo "mprime installed" ;
echo ;

echo "Please read the README file" ;

Save the file as "installmprime.sh" with out quotes in your home folder.

Type:

bash$

chmod +x installmprime.sh

bash$

sh installmprime.sh

You will be prompted to enter your 'sudo' password. This is generally the same as your login password. If you don't use sudo please run this script when logged in as root.

You may now begin stress testing by typing:

bash$ mprime

and choosing menu option 15.

Click HERE for more information on using mprime.

To remove mprime simply type:

bash$ sudo rm -rf /usr/bin/mprime

Clear downloaded packages in LInux

2010-08-30T23:34:00.002+01:00

After a while of using linux you can end up with a lot of downloaded but not installed packages and/or unwanted dependences, which can take up a lot of disk space.

It is quite easy to clean these up however.

In Debian based systems that use the apt package manager you can simply type:

bash$ sudo apt-get autoremove ; sudo apt-get autoclean

In arch linux using pacman to clean the package cache of packages that are not currently installed:

bash$ pacman -Sc

And to clean the entire package cache:

(Warning: Only do this when certain that Downgrading Packages will not be a necessity, as pacman -Scc removes all packages from the cache.)

bash$ pacman -Scc

Some programs such as Ubuntu Tweak are being developed to help with modifying and generally cleaning up the system so keep an eye out for them.

Use your webcam in a screen cast.

2010-08-30T23:12:00.002+01:00

I have seen a lot of people making linux screencasts and tutorials for linux and some have a live webcam in the corner of the screen which I feel is a nice personal touch when giving people advice.

However there is little or no documentation on how to do this so many people just leave cheese (a webcam program similar to photo-booth in mac) open in the corner but this takes up excess space. Some people advise you to do it through mplayer using a command but this takes time to work out the correct parameters for your webcam.

There is an easier way.

Gmplayer (Gnome Mplayer) has an option to do just this.

In Gmplayer you can simply go to "File">"TV">"Open Analogue TV" and it will usually display your webcam.

There are of course other media players that can do the same. VLC player for instance can use and record a webcam via the capture device menu but in my experience G-Mplayer works best.

F.lux . Better lighting for your computer inside linux.

2010-08-29T16:27:00.003+01:00

A friend pointed me towards F.Lux to change lighting on my screen depending on what time of day it is. When I had a look F.Lux comes as a tarball so I decided it would be nice to write some instructions on how to install it with a simple script.

Firstly open your editor and paste the following into it:

echo "Downloading F.lux . Please wait........" ;
wget -c https://secure.herf.org/flux/xflux.tgz ;
echo "Installing F.Lux ............." ;
tar  -xvzf xflux.tgz ;
rm -rf xflux.tgz ;
sudo  cp xflux /usr/bin/ ;
sudo chmod 755 /usr/bin/xflux ;
echo "F.lux installed" ;
echo "Please run 'xflux -l lat lon' replacing lat and lon with your latitude and longitude"

Save the file as "installflux.sh" with out quotes in your home folder.

Now open your shell, Applications>Accessories>Terminal for ubuntu users.

Type:

chmod +x installflux.sh
sh installflux.sh

You will be prompted to enter your 'sudo' password. This is generally the same as your login password. If you don't use sudo please run this script when logged in as root.

Once the script is finished you can remove the script and xflux from your home folder.

Now you can run flux with xflux command, you can also add it to your start up, But first you have to define your zipcode or latitude.Example :

xflux -l 33.55, 44.55

When i am posting this blog post, the f.lux does not support latitude over 60.

Or you can define zipcode -z instead of -l

Now add it to your startup :

System > Preferences > Sessions
(for ubuntu users)

for arch users edit your .xinitrc .

Name : Xflux
Command : xflux -l 33.55, 44.55 (DO NOT FORGET TO CHANGE the latitude or Zip!)
Comment : F.lux , Better lighting For Your Computer

Hope that helped.

Unlock the root account in sudo based systems.

2010-08-29T16:19:00.002+01:00

A lot of linux operating systems now do not have a root account. Instead you do everything through sudo which can become irritating to someone used to the power of the root.

To unlock the root account in a sudo based system, log in as a normal user, open a terminal and type

bash$ sudo passwd root

enter your sudo password and then follow the prompts to set a root password.

After this you can log in as root in a terminal although you may not be able to log in graphically as root as some login managers prevent it.

If you wish to occasionally use the root account but do not want to unlock it you can always type

bash$ sudo su -

whenever you need to log in as root. This will let you use your sudo password to log into the root account.

If you have unlocked the root account and wish to lock it again simply type

bash$ sudo passwd -l root

Reset forgotten root password

2010-08-29T16:13:00.003+01:00

Several linux based operating systems now do not have a root password and instead use sudo as a standard user. This is not for the purpose of unlocking the root account. This method is simply to reset or change a forgotten root password.

I am not liable for any damage you may incur to your system when using this method. I also do not condone the use of this method for malicious or illegal purposes.

When your computer boots up you usually get some sort of boot loader.

With GNU GRUB (very common) you usually get the option to choose an operating system or the option of pressing the Esc key to get to the grub menu.

Once you are at the Grub menu please find the newest kernel. (usually the top entry).

Press the e key to edit the entry.
Now select the second line down (the line starting with the word kernel) that looks similar to:

kernel /boot/vmlinuz-2.6.15-26-386 root=/dev/hda1 ro quiet splash

Press the e key to edit this entry.
Type the letter S (or word Single) at the end of the (kernel) line. *
Press the ENTER key
Now press the b key to boot the Linux kernel into single user mode.

At the root prompt type

bash$ passwd

to reset password and follow the prompts.

When you type your password it will seem as if you are not typing.
This is normal as * shows how many characters you have so a blank is a security feature.

Once the password has been set you can type

bash$ reboot

to restart the computer.

The root password will now be the one you changed it to.

*(If "s" or "single" do not work you can try typing "init=/bin/bash" instead)

To stop people from using this method to break into your machine you could set a grub password or a BIOS boot password.

Reset bad compizconfig settings in Ubuntu

2010-08-29T16:03:00.002+01:00

Today I saw a post on the ubuntu forums where someone had installed compiz-config-settings-manager and had made their system un-usable ie they could not log in because of bad settings and the GDM was causing problems.

After I helped him to fix it he asked me to put it somewhere accessible so others could do the same.

Only use this if your problem is caused by compiz and you need to reset it.

If it is possible to get to the GDM (Gnome display manager....... or the login screen ) then use method 1.
If you can't see the login screen at all go to method 2.

Method 1.
This should be simple enough.

Press Ctrl+Alt+F1 when at the login. You will get a terminal in fullscreen.
DO NOT BE SCARED.

(You may want to print these instructions before doing this)

Login with your username and password.
When you type your password it will seem as if you are not typing.
This is normal as * shows how many characters you have so a blank is a security feature.

You then need to stop X. ( the graphics)

Code:

sudo service gdm stop

You will be asked for your password so type it in. Again you will not see typing.

you then need to reset compiz.

Code:

gconftool-2 --recursive-unset /apps/compiz

hope that works for you.

You can then restart your graphical display.

Code:

sudo service gdm start

use Ctrl+alt+F7 to get back to your gui.

Method 2.

If you can't get as far as the login screen due to graphics issues caused by compiz config the use this method.

As your computer boots up interrupt at GRUB.

In versions of ubuntu 9.04 and before you will see a prompt on boot saying click ESC to enter grub , do so.

In 9.10 and newer you will not see this so hold the Shift key during boot and you should get to grub. When there choose the newest recovery option.

Once you boot up in recovery mode you will be presented with a small box with some options. Select 'root'.

You now have a root access command line.

You then type:

Code:

su yourusername

( obviously replacing 'yourusername' with your login name) to login as you. You then run

Code:

gconftool-2 --recursive-unset /apps/compiz

You will not have to stop or start the gdm though. When you have run the commands instead of

Code:

sudo service gdm start

type

Code:

sudo shutdown -r now

Your computer should restart and your compiz settings will be reset.

The purpose of this blog

2010-08-29T15:59:00.002+01:00

Whenever I help someone out on the forums or get help with a problem myself I often find that the help given could be useful to someone else and I always feel bad that I don't share this information enough.

That is why I have started this blog. It will be simply a log of the help I give to someone with a linux based problem.

I myself use Arch linux but have used many other distributions in the past such as Ubuntu, Debian, Fedora, Red Hat, Suse and several others.

Occasionally I may also produce guides for doing specific tasks, for instance setting up certain operating systems.

If you have any questions about my blog feel free to email me on my public address of matux.blog@gmail.com. Please do not use this as a help line or an alternative to forums.

I am doing this for no monetary gain and simply to help others.