Quickly Check Temperature Values of Hardware Components in Ubuntu

So, I missed posting yesterday. Hope this doesn’t repeat.

Today I am sharing a small bash script I wrote to check the temperatures recorded by various sensors in my laptop. Nothing incredibly smart here, just a quick but useful hack.

I am one of those unhappy Linux users who suffer from lack of driver support for their hardware. Due to some weird kernel bug or messy graphics driver, which led to incredibly high temperatures on my laptop,  I spent about a year using Linux as a VirtualBox guest in Windows; this was before Ubuntu 12.04 got shipped. During those times, my laptop used to shutdown automatically after reaching critical temperatures (100° C!) on simple tasks like watching a HD video on VLC.

What all do we need? In *buntu systems, install sensors and hddtemp tools. I am using an ATi Radeon card and proprietary driver ships with a utility for reporting temperature for the same. You can modify the script to work with nVidia cards accordingly.

sudo apt-get install lm-sensors hddtemp

Next, you need to run sensors-detect to let sensors identify all the hardware monitoring sensors present in your system.

sudo sensors-detect

Press enter to accept default options when asked.

Here is the script; hddtemp requires sudo making this script more than 3 lines:

	#!/bin/bash
	sensors
	aticonfig –odgt
	answer="N"
	if [ "$(whoami)" != "root" ]; then
	echo -ne "\n\nDo you want to know hard disk temperature (requires sudo)? (y/N) "
	read answer
	if [ "$answer" == "y" ]; then
	{
	sudo hddtemp /dev/sda
	}
	else
	{
	exit 0
	}
	fi
	else
	echo
	sudo hddtemp /dev/sda
	exit 0
	fi

view raw

temp.sh

hosted with ❤ by GitHub

I have put this script in my local bin folder for quick access. To do the same, follow the steps:

mkdir ~/bin

Put this directory in your path by putting the following line at the end of your .bashrc file (replace k4rtik by your username)

export PATH=$PATH:/home/k4rtik/bin

mv temp.sh ~/bin/temp

chmod +x temp

Now either logout and login or issue the following command to be able to access the script by just entering temp on your terminal.

source ~/.bashrc

Here is a sample run from my machine:

k4rtik: ~ $ temp
acpitz-virtual-0
Adapter: Virtual device
temp1:        +26.8°C  (crit = +127.0°C)
temp2:        +70.0°C  (crit = +85.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0:       +70.0°C  (high = +84.0°C, crit = +100.0°C)
Core 1:       +70.0°C  (high = +84.0°C, crit = +100.0°C)
Core 2:       +70.0°C  (high = +84.0°C, crit = +100.0°C)
Core 3:       +70.0°C  (high = +84.0°C, crit = +100.0°C)

Default Adapter - ATI Mobility Radeon HD 4670
                  Sensor 0: Temperature - 74.50 C

Do you want to know hard disk temperature (requires sudo)? (y/N) y
[sudo] password for k4rtik: 
/dev/sda: ST9500420ASG: 51°C

PS: Didn’t know earlier – embedding Github gists into WordPress is as easy as copy & pasting the URL. 🙂

Best Tool for Downloading YouTube Videos and Playlists

How many times does it happen that you stumble upon an awesome youtube video, you make up your mind to download it but fail short to find a good tool which actually works?

In this post I want to introduce one of the best tools for downloading videos from YouTube and other video platforms – youtube-dl. This command line tool is basically just a python script released by its developers in public domain and it works on any platform that supports Python 2.x or later including Linux, Windows or Mac OS X.

If you are on Ubuntu, you can directly install it by using:

k4rtik: $ sudo apt-get install youtube-dl

If not, just go to youtube-dl download page and save the script in your home directory. Then make the script executable by using the following command:

k4rtik: $ chmod +x youtube-dl

You can also consider moving the script to your /usr/local/bin directory to invoke it directly without specifying the path:

k4rtik: $ sudo mv youtube-dl /usr/local/bin

In its most basic form you can use this tool to download the highest quality of video by just supplying the url of the video as its argument, e.g.:


k4rtik: Videos $ youtube-dl http://www.youtube.com/watch?v=Rk62hRBDLGc
Setting language
Rk62hRBDLGc: Downloading video webpage
Rk62hRBDLGc: Downloading video info webpage
Rk62hRBDLGc: Extracting video information
[download] Destination: Rk62hRBDLGc.flv
[download] 100.0% of 5.38M at 88.81k/s ETA 00:00
k4rtik: Videos $


My preferred way of downloading videos is by using the -t flag, which saves the video with the title in the file name, e.g.:


k4rtik: Videos $ youtube-dl -t http://www.youtube.com/watch?v=Rk62hRBDLGc
Setting language
Rk62hRBDLGc: Downloading video webpage
Rk62hRBDLGc: Downloading video info webpage
Rk62hRBDLGc: Extracting video information
[download] Destination: Reasons_to_love_Ubuntu_12_04_LTS-Rk62hRBDLGc.flv
[download] 100.0% of 5.38M at 204.52k/s ETA 00:00
k4rtik: Videos $


In case, you don’t want to download the highest quality to save some bandwidth and time, youtube-dl offers choice of multiple formats. See its man-page to find out more, I prefer H264 videos at 480p size for youtube which I specify like this:


k4rtik: Videos $ youtube-dl -t -f 18 www.youtube.com/watch?v=02nBaaIoFWU
Setting language
02nBaaIoFWU: Downloading video webpage
02nBaaIoFWU: Downloading video info webpage
02nBaaIoFWU: Extracting video information
[download] Multipath_TCP-02nBaaIoFWU.mp4 has already been downloaded
k4rtik: Videos $

You can also use it for downloading complete playlists off youtube:

k4rtik: Videos $ youtube-dl -f 18 -t http://www.youtube.com/watch?v=128ll4yXUfY&list=PL2E1848DB88958935

Oh, and did I mention… it continues the download if interrupted due to network trouble or some other reason, just specify the same command again.

According to its man page it supports video downloads from other popular video hosting sites like Facebook, Metacafe, Vimeo, Yahoo!, YouTube, blip.tv, video.google.com and many more. Hope you enjoy using it for all your video downloading needs.

Ciao

K4rtik

This article first appeared at http://www.digimantra.com/tutorials/best-tool-for-downloading-youtube-videos-and-playlists/

Control All Computers in a Lab from a Single System

Quoting Dhandeep, our super-cool lab-admin:

now , all 70 systems in the lab can be switched on and switched off by single commands from the hostel…

Yes, that and a lot more is possible in our Software Systems Lab now. How? Read on…

The Setup

We have over 70 systems with Ubuntu 10.04 LTS installed on them. There is an administrative account (let’s call it admin for this post) and a guest (limited privilege) account on each. Needless to say, admin password is known only to admins and guest password is known to all who use the lab. All these systems are configured to be able to controlled remotely (read: OpenSSH server is installed on each).

Basic Idea

1. Log in via SSH without a password
2. Write your desired command and run it in background
3. Run the above in a loop for the lab’s subnet.

Detailed Steps

See Tips for Remote Unix Work (SSH, screen, and VNC) for the first step (and for more immensely useful tips on remote usage of *NIX systems).

For Step 2, here is one example command:

ssh -t admin@labsystem "echo  | sudo -S shutdown -h now" &

In the above command labsystem is usually replaced with an IP address like 192.168.xxx.xxx and the <pass> with the password of the admin account.

WARNING: it’s not suggested to use the above command out in the open to save the password from prying eyes; also note that for additional security, you need to take a measure to make sure this is not saved in bash history or if the command is in a script, it’s not accessible to others.

The requirement of ampersand at the end depends on particular usage (if you want to run, let’s say,  uptime command over ssh, you would not want the output to go to background, or you can redirect the output to some file). Putting the process in background, in this case, will help in the next step.

The -S switch for sudo makes it possible to supply the password via stdin (we had discovered this switch from sudo’s man page, but didn’t manage to conclude “echo pass |” will do the trick until we discovered it at StackOverflow)

Step 3: use your favorite scripting language (bash, python, etc.) and run the above command for all the systems of your lab subnet. An example in bash:

for ip in {101..180}
do
	ssh -t admin@192.168.xxx.$ip "echo  | sudo -S shutdown -h now" &
done

The above code snippet will run the desired command for all systems in subnet within the IP range 192.168.xxx.101 to 192.168.xxx.180. Now, you can clearly see how putting the process in the background will help – the next iteration of the loop need not wait for the command in previous iteration to finish!

In the passing, here’s a small video I shot featuring Dhandeep when he got all excited to see this working:

That’s it. Try this out, share your tricks and have some *NIX fun in your lab. 🙂

PS: I have not covered how systems can be switched on with this setup. It basically involves broadcasting a magic packet to the subnet. Hope Dhandeep comes up with a blog post on that soon. 😉 Here it is: On the push of a button..

Ciao

Kartik

Configuring Apache for Developing Multiple Websites under Ubuntu Linux

Are you a beginner in web development with LAMP environment, or a CS/IT student overwhelmed with setting up LAMP and configuring it for your web-related academic project? If yes, then you are at the right place.

We have discussed earlier how easy it is to setup a web development environment on your Ubuntu desktop. In this article we will try to have a more customized setup of this environment for easily working with multiple websites. Also find below introduction to some useful Apache utilities and configuration files on Ubuntu/Debian operating system.

One of the ways to work on multiple sites is to put different directories under /var/www and access them as http://localhost/site1, http://localhost/site2 etc. but there is another and more elegant way. We can access our sites simply by pointing our browsers to http://site1/ and http://site2 etc. To find out how, follow the tutorial.

Let’s keep all our development websites under the home directory in public_html folder. Apart from being easy to manage your website from inside your home folder, this has the benefit that you no longer need to use sudo to put files in your document root (/var/www earlier).

Open the Terminal and follow the steps for a test setup, you may customize it your preferences later:

kartik@PlatiniumLight ~ $ mkdir public_html
kartik@PlatiniumLight ~ $ cd public_html/
kartik@PlatiniumLight ~/public_html $ mkdir site1
kartik@PlatiniumLight ~/public_html $ mkdir site2
kartik@PlatiniumLight ~/public_html $ 

Create files called index.html with some content inside each of these directories to help you identify them when you walk through this tutorial:

kartik@PlatiniumLight ~/public_html $ cd site1
kartik@PlatiniumLight ~/public_html/site1 $ gedit index.html

Screenshot - index.html (~/public_html/site1) - gedit

Enter some text such as “This is site 1” and save the file. Similarly put “This is site 2” in the index.html of site2 directory.

kartik@PlatiniumLight ~/public_html/site1 $ cd ../site2
kartik@PlatiniumLight ~/public_html/site2 $ gedit index.html

Screenshot - index.html (~/public_html/site2) - gedit

Now begins the main step of the process. We have to create separate apache configuration file for each of the websites. This is fairly easy to do. We copy the default config file and modify it to our needs:

kartik@PlatiniumLight ~/public_html/site2 $ cd /etc/apache2/sites-available/
kartik@PlatiniumLight /etc/apache2/sites-available $ sudo cp default site1
kartik@PlatiniumLight /etc/apache2/sites-available $ sudo cp default site2
kartik@PlatiniumLight /etc/apache2/sites-available $ sudo vim site1

Note that I am using vim editor for making the changes as I prefer it for its amazing syntax highlight support even for various linux configuration files, you may use gedit, or any other editor instead.

We need to change the path of the DocumentRoot and Directory to the particular website’s directory, and add a ServerName directive:

ServerName site1
DocumentRoot /home/kartik/public_html/site1
<Directory /home/kartik/public_html/site1/>

Take notice of the ‘/’ near the end of the second line, it is important. Also don’t forget to add the line with ServerName directive just before the DocumentRoot directive. The lines which need to be edited are highlighted with red arrows in the following screenshot:

Screenshot - Terminal - Editing site1 config file

Save the file, similar changes need to be done for other sites also, site2 in this example.

Now we need to make entries in the hosts file so that Apache can recognize the new sites.

 kartik@PlatiniumLight ~ $ sudo gedit /etc/hosts

Add the names of the new sites after localhost as shown:

Screenshot - hosts (/etc) - gedit

Now enable the new websites by using a2ensite command:

kartik@PlatiniumLight ~ $ sudo a2ensite site1
Enabling site site1.
Run '/etc/init.d/apache2 reload' to activate new configuration!
kartik@PlatiniumLight ~ $ 

Do similarly for site2, and then run the following command to activate the new configuration:

kartik@PlatiniumLight ~ $ sudo service apache2 reload
* Reloading web server config apache2
apache2: Could not reliably determine the server's fully qualified domain name, using 127.0.1.1 for ServerName
[ OK ]
kartik@PlatiniumLight ~ $ 

You can ignore the warning which is shown in the above output. Now open your favorite browser and open http://site1 and http://site2 in different tabs, you will find the content of their respective index.html files.

Screenshot - Chromium showing both the sites

And now you are done with setting up multiple sites in Apache. You can create as many sites as desired by following the same procedure, creating a config file for each website.

Here follows an introduction to some useful Apache utilities and configuration files which aid you in this kind of a setup:

a2ensite – As we used this above, you already know what it does – enable a website among those available under /etc/apache2/sites-available directory.

a2dissite – does exactly the opposite – disable a website so that it is no longer accessible from your web server. E.g.

kartik@PlatiniumLight ~ $ sudo a2dissite site2
[sudo] password for kartik:
Site site2 disabled.
Run '/etc/init.d/apache2 reload' to activate new configuration!
kartik@PlatiniumLight ~ $ sudo service apache2 reload
* Reloading web server config apache2
apache2: Could not reliably determine the server's fully qualified domain name, using 127.0.1.1 for ServerName
[ OK ]
kartik@PlatiniumLight ~ $ 

This will disable site2 from loading any more. You can use a2ensite to enable it again.

Similar to sites, there are corresponding directories and commands for controlling apache modules also. Available modules are kept under /etc/apache2/mods-available directory and enabled ones are symlinked under mods-enabled directory. The corresponding commands are – a2enmod and a2dismod. E.g.

kartik@PlatiniumLight /etc/apache2/conf.d $ sudo a2enmod rewrite
Enabling module rewrite.
Run '/etc/init.d/apache2 restart' to activate new configuration!
kartik@PlatiniumLight /etc/apache2/conf.d $ sudo service apache2 reload
* Reloading web server config apache2
apache2: Could not reliably determine the server's fully qualified domain name, using 127.0.1.1 for ServerName
[ OK ]
kartik@PlatiniumLight /etc/apache2/conf.d $ 

This will enable Apache rewrite module which is required my some CMS solutions like Drupal to enable pretty URLs.

Hope this article was useful to you, feel free to ask your doubts, problems you face, or any similar tips you might have through comments.

An edited version of this article first appeared at http://www.muktware.com/articles/1876

Remove Orphaned Actions in Drupal 7

I recently discovered this weird behaviour by drush when I disabled Comment module in Drupal 7 – drush was giving the following warning whenever I tried enabling/disabling any module:

WD actions: 3 orphaned actions (comment_publish_action, comment_save_action, comment_unpublish_action) exist in the actions table. Remove orphaned actions

After some scrounging on drupal.org (where most solutions were meant for Drupal 6), I found the solution at http://blog.devkinetic.com/node/9 Just execute the following once:

drush php-eval "actions_synchronize(actions_list(), TRUE);"

Easiest Way to Setup a Web Development Environment on Ubuntu-based Distros

Did you know how easy it is to get a basic web development environment on your Ubuntu-based Linux distribution? Guess what, it just takes 2 commands on the Terminal:

sudo apt-get install tasksel

This will install a small utility which lets you install a lot packages grouped together as software collections.

sudo tasksel

Launch tasksel and select ‘LAMP server‘ by pressing the SPACE key, press ENTER when you are done (see attached screenshot). It will take some time for the required packages to download and install. Near the end of setup, the installer will ask you to create a password for MySQL‘s root user.

Select LAMP Server among the choices in Tasksel

After the installer finishes, you have the environment ready. Head over to your favorite browser and open http://localhost If everything went fine, the page will say It works!

It works!

Now you can start creating websites by putting your html, php, etc. files under /var/www directory or just choose to go with CMS solutions like Drupal, WordPress or Joomla.

The author also recommends to install phpmyadmin package if you happen to work with MySQL databases.

An edited version of this article first appeared at http://www.muktware.com/articles/08/2011/1348

Use rfkill to Enable/Disable Wireless on your Linux Laptop

This notebook computer is connected to a wireless access point using a PC card wireless card. (Photo credit: Wikipedia)

Imagine a situation when you have to book an air/train ticket in a jiffy, or check an important mail quickly and the only option you have is a wi-fi connection from either your smart phone or surroundings, and you have only some Linux variant installed on your system. And even after installing all the necessary drivers, you are unable to get the wi-fi on your laptop working? Frustrating right? If yes, then you might want to read on about this useful utility called rfkill which you can keep in handy for those weary situations.

I own a Dell Studio XPS 1645 and have always found it cumbersome to get the wi-fi working on my system, mainly during those geek/hacker meetups, the only times I have to use wireless Internet. I remember randomly switching wireless on and off through the hardware switch and rebooting my system multiple times in order to get it working. Well, this was the situation until I discovered rfkill – a tool for enabling and disabling wireless devices including Wireless LAN, Bluetooth, etc. Here follows a tutorial on how to use it (fire up the Terminal before proceeding):

rfkill’s list command lets you see all the available devices, if you don’t find any of your devices, make sure you have turned the hardware switch ON and have installed the drivers for each. Here is what I get on my system after enabling the hardware switch:

kartik@PlatiniumLight ~ $ rfkill list
0: brcmwl-0: Wireless LAN
Soft blocked: yes
Hard blocked: no
1: dell-wifi: Wireless LAN
Soft blocked: yes
Hard blocked: no
2: dell-wwan: Wireless WAN
Soft blocked: yes
Hard blocked: yes
3: hci0: Bluetooth
Soft blocked: no
Hard blocked: no
kartik@PlatiniumLight ~ $

Apart from Bluetooth, I usually find all other devices to be in a random state of yes or no. To enable them, issue the unblock command as shown:

kartik@PlatiniumLight ~ $ rfkill unblock 0
kartik@PlatiniumLight ~ $ rfkill unblock 1
kartik@PlatiniumLight ~ $ rfkill unblock 2
kartik@PlatiniumLight ~ $ rfkill list
0: brcmwl-0: Wireless LAN
Soft blocked: no
Hard blocked: yes
1: dell-wifi: Wireless LAN
Soft blocked: no
Hard blocked: no
2: dell-wwan: Wireless WAN
Soft blocked: no
Hard blocked: no
3: hci0: Bluetooth
Soft blocked: no
Hard blocked: no
kartik@PlatiniumLight ~ $

You can also try using the unblock all command for enabling all the devices together:

kartik@PlatiniumLight ~ $ rfkill unblock all

Sometimes it happens that even after unblocking once, some device(s) may show up as blocked (see the 0th device above, which shows hard blocked as yes). To correct this just issue the unblock command again for that particular device:

kartik@PlatiniumLight ~ $ rfkill unblock 0
kartik@PlatiniumLight ~ $ rfkill list
0: brcmwl-0: Wireless LAN
Soft blocked: no
Hard blocked: no
1: dell-wifi: Wireless LAN
Soft blocked: no
Hard blocked: no
2: dell-wwan: Wireless WAN
Soft blocked: no
Hard blocked: no
3: hci0: Bluetooth
Soft blocked: no
Hard blocked: no
kartik@PlatiniumLight ~ $

When you get all the devices unblocked, you will not face any trouble connecting to wi-fi devices around. 🙂

Bonus Tip: If you have a common hardware switch for wireless radios, you can turn off additional devices like Bluetooth (or vice versa) to save some battery life using the block command of rfkill:

kartik@PlatiniumLight ~ $ rfkill block 3
kartik@PlatiniumLight ~ $ rfkill list
0: dell-wifi: Wireless LAN
Soft blocked: no
Hard blocked: no
1: dell-wwan: Wireless WAN
Soft blocked: no
Hard blocked: no
2: brcmwl-0: Wireless LAN
Soft blocked: no
Hard blocked: no
3: hci0: Bluetooth
Soft blocked: yes
Hard blocked: no
kartik@PlatiniumLight ~ $

Stay Dignified!

• Kartik

Originally published at http://www.digimantra.com/linux/rfkill-enabledisable-wireless-linux-laptop/

Restore GRUB in Ubuntu after installing Windows

So, you had Ubuntu (or any of its derivative) installed on your system and then decided to install Windowswith it? And after installing Windows, you can’t load Ubuntu again, just follow the steps to get this working:

1. Boot the system using a Ubuntu Live disk.
2. After you reach the desktop, open Terminal (press Alt+F2, type “gnome-terminal” without quotes and press enter).
3. Execute:

   sudo fdisk -l

4. Execute (after replacing sda7 with your corresponding linux root partition, which you will be able to identify from the output of the previous command – it’s usually marked as Linux):

   sudo mount /dev/sda7 /mnt

5. Just check whether you got the root partition right, if you have more than 1 linux partition on your hard disk:

   cd /mnt
ls

   If the listing contains names like bin, boot, dev, etc, home, tmp, usr, var, et cetera then you are in the right partition. If not, unmount this using:

   sudo umount /mnt

   and follow the above step again using another partition.

6. Now:

   sudo grub-install --root-directory=/mnt /dev/sda

   and grub will get installed if it shows no errors.

7. As a last step unmount your /mnt directory using:

   sudo umount /mnt

   and restart the system. You will find the friendly grub boot manager giving you the choice to boot into Linux or Windows.

An edited version of this post first appeared at http://www.digimantra.com/linux/restore-grub-in-ubuntu-after-installing-windows/

—
Kartik – The TechGlider Guy
http://www.techglider.in

Input/Output using Text Files Without Diving into File Handling

It’s quite easy to use text files as input source or to redirect the output of programs to files if you happen to do programming on a unix/linux terminal.

Let’s take an example of a usual random number generator:

// random.c
#include <stdio.h>
#include <stdlib.h>
int main ()
{
        int i, count=99999999;
        for (i = 0; i < count; i += 1)
                printf("%d\n", (int)random());

        return 0;
}

Let’s generate an executable by the name random.out for this program:

gcc random.c -o random.out

Then, to redirect it’s output we can use output redirection offered by the bash shell (tip: reduce the value of count in the code above if you don’t want to generate a 1 GiB file and waste both time and disk space):

./random.out > randout.txt

The > (greater than symbol) redirects the standard output to the file named randout.txt.

To use a file as input source, the < (less than symbol) is used, e.g.:
./quicksort.out < randout.txt

You can also combine both redirections:
./quicksort.out < randout.txt > sortedfile.txt

Another nifty concept is pipelining (or piping), which can be used to redirect output from a program as input of another, for example, to generate random numbers on fly and feed them to my quicksort program I can do:
./random.out | ./quicksort.out > sortedfile.txt

In the above, input source files is not specified as required number of inputs are taken directly from random program. | (vertical pipeline symbol) enables this functionality.

The program which you create for using these text files should just be able to receive the input and return the output separated by newline characters. Using a for loop for taking definite number of inputs is preferred.

See also: Redirection on Wikipedia

Don’t hesitate to leave a comment if this post helped you.

EDIT (2011-02-24): Got into trouble while using this method for taking input in my DSA lab last week. Apparently this works only if the for loop for taking input does just that – input iteratively in an array – and nothing else. While calling insert function in a binary search tree implementation along with the scanf inside the input loop lead to weird input to the program. File handling seems to be the only solution in this situation. Though if you find any explanation for this behavior of bash shell, please do share in the comments.

Accessing NIT Calicut’s Ubuntu Mirror (in the campus)

As the FOSSCell webserver has been facing troubles these days, the instructions to use the Ubuntu Mirror of NIT Calicut that were available on the FOSSCell wiki are being missed by many. Here they are though not in the informative way as were given on the wiki:

Add the following to your /etc/apt/sources.list and comment out other similar repos.


deb http://192.168.40.97/ubuntu lucid-updates main multiverse restricted universe
deb http://192.168.40.97/ubuntu lucid-security main multiverse restricted universe
deb http://192.168.40.97/ubuntu lucid main multiverse restricted universe


Then update the repo database using:
sudo apt-get update