Linux: Difference between revisions

From 太極
Jump to navigation Jump to search
Line 1,236: Line 1,236:
=== Monitor harddisk health by command line ===
=== Monitor harddisk health by command line ===
Using [https://www.howtoforge.com/tutorial/monitor-harddisk-with-smartmon-on-ubuntu/ smartctl].
Using [https://www.howtoforge.com/tutorial/monitor-harddisk-with-smartmon-on-ubuntu/ smartctl].
=== Troubleshoot and repair network problems ===
http://www.linuxuser.co.uk/features/troubleshot-repair-linux-networks


=== Monitor network by command line ===
=== Monitor network by command line ===

Revision as of 13:21, 2 July 2015

Handy Linux tips

Beautiful desktop

Virtual consoles

Linux allows virtual consoles (aka virtual terminals) to be opened while an X Window System is executing.

Use Ctrl+Alt+F1 to open a virtual console-- there are six virtual text-based consoles (F1 to F6). Use Ctrl+Alt+F7 to return to the X Window System.

How much resource is used by a process

Find the process ID first by ps -ef | grep APPLICATIONAME. Then 

ps -p <pid> -o %cpu,%mem,cmd

For example, 

$ ps -ef | grep akregator
brb      15013  1942  1 10:41 ?        00:00:05 akregator --icon akregator -caption Akregator
brb      15186 24045  0 10:50 pts/11   00:00:00 grep --color=auto akregator
$ ps -p 15013 -o %cpu,%mem,cmd
%CPU %MEM CMD
 1.0  0.8 akregator --icon akregator -caption Akregator

Query whether the OS is 64-bit or 32-bit

SYSTEM_ARCH=getconf LONG_BIT
echo $SYSTEM_ARCH

Switch user in command line

use 

su newusername

to switch to another user.

Find binary file location

  • which - Display the full path of shell commands. See examples from cyberciti.biz.
$ which ls
/bin/ls
  • whereis - locate the binary, source, and manual page files for a command. See examples from cyberciti.biz.
$ whereis ls
ls: /bin/ls /usr/share/man/man1p/ls.1p.gz /usr/share/man/man1/ls.1.gz
  • type -a
$ type -a ls
ls is aliased to `ls --color=tty'
ls is /bin/ls

Use locate command mindfully. It is used to find the location of files and directories. Note that locate does not search the files on disk rather it searches for file paths in a database. For example, the following command will search .png files over the system (not only the personal directory). 

locate "*.png"

IP address fundamental

http://www.howtogeek.com/133943/geek-school-learning-windows-7-ip-addressing-fundamentals/

Change to root shell

sudo su -

This can be useful when running 'su' or 'su -' failed because of authentication failure.

See also

For sudo to work, my account ('debian' in this case) has to be included in the config file /etc/sudoers. 

debian  ALL=NOPASSWD: ALL

When sudo is invoked, it asks for the password of the user who started it.

Bash shell programming

http://bash.cyberciti.biz/guide/Main_Page

Redirect standard error

http://bash.cyberciti.biz/guide/Standard_error. Use 2> operator. 

command 2> errors.txt

ls | more without lose color

ls --color | more

Most likely your ls is aliased to ls --color=auto. If you do ls --color (which is morally equivalent to ls --color=always), that will force it to turn on colors.

ls output with color background

stackexchange or askubuntu.

In my case, after I apply chmod 755 -R XXXX, the weird green background color goes away.

colordiff-color in terminal diff

PS. For a GUI version of diff, Meld works fine. Need to install first. apt-get install colordiff. http://www.cyberciti.biz/programming/color-terminal-highlighter-for-diff-files/ 

sudo apt-get install colordiff
diff -y file1 file2 | colordiff
# Ignore same rows (two ways):
# diff -C0 file1 file2 | colordiff
# diff -U0 file1 file2 | colordiff

where -y option means to show the output in two columns.

Colordiff.png

Interpretation of the diff output:

The first line of the diff output will contain:

  • line numbers corresponding to the first file,
  • a letter (a for add, c for change, or d for delete), and
  • line numbers corresponding to the second file.

In our output above, 2,4c2,4 means: "Lines 2 through 4 in the first file need to be changed in order to match lines 2 through 4 in the second file." It then tells us what those lines are in each file:

  • Lines preceded by a < are lines from the first file;
  • lines preceded by > are lines from the second file.
  • The three dashes ("---") merely separate the lines of file 1 and file 2.
2,4c2,4
< I need to run the laundry.
< I need to wash the dog.
< I need to get the car detailed.
---
> I need to do the laundry.
> I need to wash the car.
> I need to get the dog detailed.

Run diff with large files

Meld freezes When I tested it with two large files (800k & 936k lines coming from human gtf files). Actually the whole linux system became unresponsive.

Print certain rows/lines of a text file

The following example will print out lines 10 to 60 of FILENAME. 

sed -n '10,60p' FILENAME

Or to print out line 60, 

sed -n '60p' FILENAME

It seems this method is not as fast as I expected. For example, the tail command will immediately print out the result without waiting!

output colored console to html

Use ansi2html.sh. It only requires gawk.

  1. Use wget to download it
  2. sudo apt-get install gawk
  3. chmod +x ansi2html.sh
  4. colordiff file1 file2 | ./ansi2html.sh > diff.html

using a the result of a diff in a if statement

ls -lR $dir > a
ls -lR $dir > b

DIFF=$(diff a b) 
if [ "$DIFF" != "" ] 
then
    echo "The directory was modified"
fi

Another example 

if [ "$(diff file1.html file2.html)" == "" ]; then echo Same; else echo Different; fi

Listen to pandora in Europe: install squid proxy

http://www.cyberciti.biz/faq/access-pandora-radio-using-proxy-server-outside-usa/

Interestingly, the firefox connection settings should choose HTTP Proxy instead of 'SOCKS host'.

Bypass blocked website from work computer

ssh -D 4096 user@remoteip
ssh -D 4096 -p 23 user@remoteip

This will require you to enter the password and leave you in the remote machine. If a nonstandard port is required, we can use -p option.

Now in the firefox, we need to go to Edit -> Preferences -> Advanced -> Network tab -> Settings... Check 'Manual proxy configuration' (The default is 'Use system proxy settings') and enter 'localhost' for SOCKS Host and '4096' for the Port. Don't enter 'localhost' in the HTTP Proxy.

On Windows, we can use Putty. In short, in the left-hand panel, navigate through Connection > SSH > Tunnels. Enter 4096 in the Source Port box and select the Dynamic radio button. Click Add and “D4096″ will appear in the Forwarded Ports list. The setting in the firefox end is the same. See also my Windows wiki page.

Linux journal also put a video on youtube. We can use http://www.ipligence.com/geolocation to check the current location. The port number is 1080 in the example. The example actually also use '-N' option which means no interaction; i.e. ssh -N -D 1080 user@remoteip. So we won't see anything after we type our password. Once we want to stop SOCK proxy, we just need to hit Ctr+C on terminal.

Graphical way to display disk usage

For example, to use xdiskusage, we run apt-get install xdiskusage and launch it by xdiskusage ~/.

  • Ubuntu has a built-in program called "Disk Usage Analyzer". Just search it from Dash. Looks useful!

Display files sorted by modified date in a directory recursively

stat --printf="%y %n\n" $(ls -tr $(find DIRNAME -type f))

find -type f -printf '%T+\t%p\n' | sort -n

Both of methods give the same output. Note the latest changed file is shown at the bottom of the output.

Display directory size with sorting and human readable

Use ncdu program (more interactive). Although it is a command line program, we can use the mouse to move through each directory to see its sub-directories. 

sudo apt-get install ncdu
ncdu

And the du method. 

du -h ~/ --max-depth=1
du -h ~/ --max-depth=1 --exclude ".*" | sort -nr | cut -f2 | xargs -d '\n' du -sh

The --exclude is to hide hidden directories.

Note that the 'du' commands may be cheating. See the following screenshot.

DiskUsage.png

The discrepancy is explained by 'sector'. See http://askubuntu.com/questions/122091/difference-between-filesize-and-size-on-disk. 

$ sudo dumpe2fs /dev/sda1 | grep -i "block size"
dumpe2fs 1.41.14 (22-Dec-2010)
Block size:               4096

To show a file size in terms of blocks, we can use 

ls -s

So for example, if a file takes 150 blocks, and if a block takes 4096 bytes, then the file takes 150*4096/1024 KB on disk.

Apache benchmark (ab) testing

ab -n 100 -c 10 http://taichi.selfip.net:81/
ab -n 100 -c 10 http://taichi.selfip.org/

rsync with exclude files/directories

See http://www.thegeekstuff.com/2011/01/rsync-exclude-files-and-folders/. The key is excluded files are relative to the current directory even we specify the absolute path. For example /path1/path2/file does not mean the file is located under /path1/path2; it means the file is located under ./path1/path2. 

rsync -avz --exclude '/path1/path2/file' source/ destination/

rsync with -a option

The -a flag in there stands for “archive,” and it’s important to include. It makes sure that the sync command is recursive (meaning any sub-folders and files inside of old_movies are copied too) and it’s important for preserving all of those modification dates, symbolic links, permissions, and other goodies we talked about earlier.

rsync with non-standard port

Use -e option 

rsync -avz -e "ssh -p 23" mydir user@remoteip:

rsync with progress bar

Use --progress option. 

rsync -avz --progress file1 file2

The 'rsync' command works on transferring files local to local too.

Or it is better to use -P option which is the same as --partial --progress. When it is used you’ll get a progress dialog at the command line that shows you which file is currently transferring, what percentage of that transfer is complete, and how many more files are left to check. As each file completes, you’ll see an ever-growing list of completed file transfers, which is great for making sure everything transfers successfully. It also allows you to easily resume suspended or interrupted transfers. Combined, you can see how it’ll show you which file was the last one to go, where it failed, and if it failed, give you the option to resume. It’s a pretty powerful combination.

rsync on Windows

Download and install command line rsync from http://www.rsync.net/resources/howto/windows_rsync.html. The website also provides a documentation. Some people are concern about the license issue. The website here provides a link to the free, old but usable version 4.0.5 which is newer than I tested v3.1.0.

Below are my note by using cwrsync v3.1.0 installer got from http://www.rsync.net. 

cd C:\Program Files (x86)\cwRsync\bin
ssh-keygen -t rsa -N ''
rsync -av "/cygdrive/c/Users/brb/.ssh/id_rsa.pub" [email protected]:.ssh/authorized_keys

rsync -av "/cygdrive/c/Users/brb/Downloads/cytokineMC.txt" [email protected]:Downloads/

scp with non-standard port

Use -P argument. 

ssh -P 23 myfile user@remoteip:

scp or ssh without password

http://www.thegeekstuff.com/2008/06/perform-ssh-and-scp-without-entering-password-on-openssh/

  1. Verify that local-host and remote-host is running openSSH
  2. Generate key-pair on the local-host using ssh-keygen
  3. Install public key on the remote-host
  4. Give appropriate permission to the .ssh directory on the remote-host
  5. Login from the local-host to remote-host using the SSH key authentication to verify whether it works properly
  6. Start the SSH Agent on local-host to perform ssh and scp without having to enter the passphrase several times
  7. Load the private key to the SSH agent on the local-host
  8. Perform SSH or SCP to remote-home from local-host without entering the password

Another option is to use ssh -i IDENTITY_FILE. See superuser.com.

ssh key

SSH key is useful if you want a password-less login to a remote system. Some useful resources:

The steps are

  • Check if there is an existing key
ls -al ~/.ssh
  • Create a new RSA key pair:
ssh-keygen -t rsa
  • Copy the public key to a remote host ([email protected]) over ssh. The current user (eg brb) and the remote user (eg git)have not any relationship (they most likely have different user names):
ssh-copy-id [email protected]

Or 

cat ~/.ssh/id_rsa.pub | ssh [email protected] "mkdir -p ~/.ssh && cat >>  ~/.ssh/authorized_keys"
  • Test if this is working by trying 'ssh [email protected]'.
  • To disable the password for root login. Type sudo nano /etc/ssh/sshd_config
PermitRootLogin without-password

Then run the following to put the changes into effect: 

reload ssh
# Or service ssh restart

If we like to ask all users to use key-based to log in, we can modify the line 

PasswordAuthentication no

in sshd_config.

Multiple ssh keys

If we want to use a specific key in ssh, use 

ssh -i ~/.ssh/xxx_id_rsa [email protected]

See also ssh-add & ssh-agent.

ssh key management

Copy ssh keys to another computer

http://askubuntu.com/questions/134975/copy-ssh-private-keys-to-another-computer

Preserve ssh keys when upgrading computers

# ls -l /etc/ssh/*key* > ~/key_list
# mkdir ~/serverkeys && cp -p /etc/ssh/*key* ~/serverkeys/
# cp -p ~/serverkeys/*key* /etc/ssh
# ls -l /etc/ssh/*key* | diff - ~/key_list

If diff produces no output, you're finished.

Handling the ssh key change when connecting to a remote machine

An article from cybercitz.biz.

  • Method 1. Remove the key using ssh-keygen -R command.
$ ssh-keygen -R {server.name.com}
$ ssh-keygen -R {ssh.server.ip.address}
$ ssh-keygen -R server.example.com
  • Method 2. Add correct host key in /home/user/.ssh/known_hosts
  • Method 3. Just delete the known_hosts file If you have only used one ssh server

ssh log files

It is also helpful to check /etc/hosts.allow and /etc/hosts.deny for any possible wrong configuration.

The device is busy

See http://ocaoimh.ie/2008/02/13/how-to-umount-when-the-device-is-busy/ 

# fuser -m /dev/sdc1
/dev/sdc1: 538
# ps auxw|grep 538
donncha 538 0.4 2.7 219212 56792 ? SLl Feb11 11:25 rhythmbox

Another handy one is: 

umount -l /dev/sdwhatever

Kill a process

# find the PID
pgrep ProgramName
# Kill the ProgramName process
kill -9 PID

Create an ext3 file system on a USB flash drive

umount /dev/sdb1   (depending on the device of course)
sudo mkfs.ext3 /dev/sdb1
sudo e2label /dev/sdb1 usbdrive   (change the label)

We can create MS-DOS file system by 

sudo mkfs.vfat /dev/sdb1

Port forwarding

For example, we like to use source port number 8080, destination port 80 (the http port), and destination server www.ubuntuforums.org. 

ssh -L 8080:www.ubuntuforums.org:80 <host>

Where <host> should be replaced by the name of your laptop. The -L option specifies local port forwarding. For the duration of the SSH session, pointing your browser at http://localhost:8080/ would send you to http://www.ubuntuforums.org/.

To stop the ssh session, use ps -ef to find the process id and kill it.

Add a new user with home directory

adduser xxx

adduser is better than useradd since useradd does not create home directory and it does not even ask the password for new user. adduser will interactively ask user information.

To delete the user and home directory, use 

deluser --remove-home xxx

Compress a folder without full path name

Suppose we want to compress the folder ~/Documents and its subfolders. We want to include Documents folder name but not /home/brb/Documents name. 

tar -czvf tmp.tar.gz -C ~/ Documents
tar -tzvf tmp.tar.gz

List contents of tar.gz

tar -tzvf myfile.tar.gz

Extract tar.xz file

tar --xz -xvf filename.tar.xz

Note the double slash in front of xz.

Extract one files from tar.gz

Extract a file called etc/default/sysstat from config.tar.gz tarball: 

$ tar -zxvf config.tar.gz etc/default/sysstat

Noe that a new directory etc/default will be created under the current directory if it does not exist.

Wildcard based extracting

You can also extract those files that match a specific globbing pattern (wildcards). For example, to extract from cbz.tar all files that begin with pic, no matter their directory prefix, you could type: 

$ tar -xf cbz.tar --wildcards --no-anchored 'pic*'

To extract all php files, enter: 

$ tar -xf cbz.tar --wildcards --no-anchored '*.php'

Show folder size for one level only

du --max-depth=1 -h

The graphical tool is called Disk Usage Analyze which is already available on Ubuntu.

Soft link

ln -s /full/path/of/original/file /full/path/of/soft/link/file

Backup

  • Use dropbox folder.

If we don't want to install dropbox software, we can install bash dropbox uploader: http://www.andreafabrizi.it/?dropbox_uploader OR https://github.com/andreafabrizi/Dropbox-Uploader

It allows to upload/download/delete/list files and show info of user. The version I am using is v0.9.7. It works on linux, Windows/Cygwin, Raspberry Pi, etc.

I install it under ~/Downloads/andreafabrizi-Dropbox-Uploader-cdc2466 directory

Instruction with screenshots: http://www.jobnix.in/dropbox-command-line-interface-cli-client/

Sample usages: 

./dropbox_uploader.sh  list /
./dropbox_uploader.sh  upload ~/Desktop/ConfigurateNote.txt

Running a cron job as a user

  • crontab cron-file-winter; crontab -l

Make sure the .sh file gives a complete path. For example, 

#!/bin/sh
R --vanilla  < arraytoolsip.R

does not work in cron job although it works perfect when we manually run it from the right path. The sh file should be 

#!/bin/sh
R --vanilla  < $HOME/Dropbox/scripts/arraytoolsip.R

Cron knows nothing about your shell; it is started by the system, so it has a minimal environment. If you want anything, you need to have that brought in yourself. For example, to use 'ifconfig' command, I need to give it a complete path in my script file. 

$ cat syncIP 
/sbin/ifconfig eth0 | grep 'inet addr:' | cut -d: -f2 | awk '{ print $1}'

and the cron job 

06 15 * * * /home/MYUSERNAME/Ubuntu\ One/syncIP > $HOME/Ubuntu\ One/ip.txt 2>&1

See here on how to add environment variable into cron environment.

Running crontab as root

Use sudo crontab -e to edit. After saving it, no need to initialize it. Use sudo crontab -l to list the cron job.

Mount drive

/etc/fstab

usb drive

Run 

fdisk -l

OR 

dmesg | grep -i "SCSI"

to confirm the USB device is detected.

Now suppose the usb device is found in dev/sdb1. 

mkdir /mnt/usb
mount -t vfat -o rw,users /dev/sdb1 /mnt/usb

The above mount command assumes the usb drive has Windows vfat partition and users give non-root users the ability to unmount the drive. If the USB drive is partitioned linux ext2/3, we can merely run mount command as 

mount /dev/sdb1 /mnt/usb

At the end, run umount command like 

umount /mnt/usb

To make the mounting automatically, edit the file /etc/fstab. 

/dev/sdb1       /mnt/usb           vfat    defaults        0       0 
/dev/sdb1       /mnt/usb           ext3    defaults        0       0

and run 

mount -a

Mount an iso file

sudo mkdir -P /mnt/mount_point # create a mount point
sudo mount -o loop /home/user/disk.iso /mnt/mount_point
mount  # verify

Mount remote Windows share

NTFS usb drive in xubuntu

http://xflinux.blogspot.com/2011/01/mount-ntfs-volumes-automatically-in.html 

sudo apt-get install ntfs-config

Now go to Applications>> System>> Ntfs Configuration Tool

Expand the "Advanced Configuration" and select all those partitions you want to be auto mounted and writable( The tool will detect all partitions at its startup).

Make sure the " Enable write support for internal devices" option is selected. Now click Close.

Many drives, one folder

run commands in a background and allow log off

nohup /path/to/script >output 2>&1 &

vi keys

  • editmode / insert: "i"
  • editmode / append: "a"
  • leave edit mode: "ESC"
  • save: ":w"
  • quit: ":q"
  • quit without saving: ":q!"
  • save and quit: ":wq"

Some helps

Cloud

http://slidedecks.wilmoore.com/2012-confoo/diy-private-cloud-using-virtualBox-and-chef/#66

http://www.datacentermap.com/blog/cloud-software-389.html

Commercial cloud services

  • DigitalOcean - simple pricing. One droplet = one server.
  • Amazon EC2
  • Google cloud
  • Microsoft Azure

U-boot

http://www.denx.de/wiki/U-Boot

Pandora linux client

COW (copy on write) file system

The cow filesystem was found on xubuntu live CD. See http://en.wikipedia.org/wiki/Copy-on-write

Apache redirection

http://cran.r-project.org/mirror-howto.html

Important linux directories

  • /bin - executables used by the base system
  • /boot
  • /dev
  • /etc - configuration files
  • /media
  • /mnt
  • /opt - optional application packages
  • /proc - process information only
  • /sbin - critical executables for running the system, but should be used by superuser
  • /usr - non-critical files. Inside is /usr/bin, which contains most of the libraries used by apps.
  • /var - variable data such as databases, mails spools and system logs.

Difference of /bin, /sbin, /usr/local/bin, ...

  • /bin : For essential binaries; e.g. bash, cat, ls.
  • /sbin : is similar to /bin but for scripts with superuser (root) privileges required; e.g. shutdown command is located here. Local users have to use sudo to run binaries here.
  • /usr/bin : Same as first, but for general system-wide & non-essential binaries; e.g. grep, zip, docker, etc.
  • /usr/sbin : Same as above, but for scripts with superuser (root) privileges required.
  • /usr/local/bin or /usr/local/sbin for system-wide available (personal) scripts.

If you want to create your own scripts and make them available to all users, you’re pretty safe adding them to /usr/local/bin. Or to add my scripts to my local bin (~/bin) and then I create a symbolic link in /usr/local/bin to the commands I want to make public. As a result, I can manage all my scripts from the same directory but still make some of them publicly available since /usr/local/bin is added to $PATH. See this post.

DHCP lease time

On Windows, it is 8 days.

Check a file's encoding

file -bi myfile

For example, 

mli@PhenomIIx6:~$ file -bi Downloads/hmv_.rc 
text/x-c++; charset=utf-16le
mli@PhenomIIx6:~$

Know you system using the command line

System monitor tools

Git and Github

git config --global credential.helper cache

My example of working on a new repository called 'network'.

  1. Follow https://help.github.com/articles/create-a-repo to create a new repository. For convenience, I also check the button to create README file.
  2. Click 'GitHub' icon on Windows Desktop. Look at the LHS and click on the word 'github'. Click 'clone' button (This can be accomplished by git clone https://github.com/arraytools/network.git). The new repository will appear under C:\Users\USERNAME\Documents\Github\ directory. Now Click 'Git Shell' icon on the Windows Desktop and go to C:\Users\USERNAME\Documents\Github\network directory where 'network' is my repository's name.
git config --global color.ui auto  # colorize the output of git
git init

git add client.c
git add server.c
git add server2.c
git commit -m 'first commit'

git status
# On branch master
# Your branch is ahead of 'origin/master' by 1 commit.

# git remote add origin https://github.com/arraytools/network.git
# git push <remote> <branch>
git push origin master

# If we modify any file, we need to go through git add, git commit and git push 3 commands.

# get commit_id
git log
# get a specific version
git checkout commit_id

# after an examination, we want to get the latest version
git pull origin master
# If you do not want to merge the upstream changes wit your local repository, run git fetch to 
# fetch the changes and then git merge to merge the changes. 
# git pull is just a combination of fetch and merge.

# To rename a file
git mv originafile newfile
git commit -m "rename orginalfile"
git push

# To remove a file
rm myfile
git add . -A
git commit -m "remove a file"
git push

# Revert to origin's master branch's version of file
# http://stackoverflow.com/questions/1817766/revert-to-origins-master-branchs-version-of-file
# 1. Assuming you did not commit the file, or add it to the index, then:
git checkout filename
# 2. Assuming you added it to the index, but did not commit it, then:
git reset HEAD filename
git checkout filename

# Compare difference in local and remote repository
git fetch    (the syntax is the same as git pull, but it doesn't automatically merge)
git status
git diff master origin/master --color --name-only (color, Show only names of changed files.)

# To force to overwrite local files on pull
git fetch --all
git reset --hard origin/master

Check the https://github.com/arraytools/network. The commit goes to the repository!

In summary: add and commit are local operations, push, pull and fetch are operations that interact with a remote.

If we want to checkout a specific commit on a new computer, we can use (here we use Qt repository as an example) 

git clone https://xxxxxx/xxxxxx/Qt.git
cd Qt       
git log --oneline
git checkout SHA1  (7 digits obtained from git log --oneline commandis enough)

After that we can run 

# move HEAD to origin
git checkout origin/master

# Visualize using text mode
git log --graph --oneline --date-order --decorate --color --all

Using GUI client: gitg 

sudo apt-get install gitg

To switch among different branch (eg. github project page is located in gh-pages branch of a repository), 

git checkout gh-pages
git checkout master

To avoid being asking the username each time when you use 'git push', you can modify the .git/config file of your local repository. This file contains a section called 'remote' with an entry called 'url'. The 'url' entry should contains the https link of repository you're talking about. When you prefix the host url with your username, git shouldn't be asking for your username anymore. Here's an example: url = https://[email protected]

Set up a new local repository

mkdir /path/to/your/project
cd /path/to/your/project
git init
git remote add origin https://[email protected]/arraytools/REPOSITORYNAME.git

git config --global user.name "YOUR NAME"
git config --global user.email "YOUR EMAIL ADDRESS"

echo "arraytools" >> contributors.txt
git add contributors.txt
git commit -m 'Initial commit with contributors'
git push -u origin master

Already has a git repository on my computer

cd /path/to/my/repo
git remote add origin https://[email protected]/arraytools/REPOSITORYNAME.git
git push -u origin --all # pushes up the repo and its refs for the first time
git push -u origin --tags # pushes up any tags

Rename a repository

  1. Go to github.com, open the project and click Settings button on the left-bottom corner. Change the repository name on top.
  2. On local machine, rename the directory. Go to the directory. Issue
git remote -v

to get the ULR for the current working copy. Suppose the url is [email protected]:someuser/someproject.git. Now issue the following command to change to the new repository 

git remote set-url origin [email protected]:someuser/newprojectname.git

Git Tips

  • git difftool FILENAME -y will launch 'meld' (if it has been installed before) to compare the file between revisions by using custom tools. It has to be run before we call git add. This is quite convenient since you can double check before running git commit. The '-y' argument is used to launch a diff tool without a prompt. See the documentation here.
  • If we want to monitor/find files that have been changed since last pull, run
git fetch origin/master
git diff --name-only origin/master
  • If we want to run git pull and also overwrite possibly changed local file, we use (see stackoverflow)
git fetch --all
git reset --hard origin/master
  • git status shows Your branch is Ahead by X commits after running git pull. See this post. The solution is to run git fetch after git pull.

Create a git server (command line only)

Follow the instruction on git-scm.com. It works. I tested it by

  1. create a git account (called 'git') on my host machine.
  2. sudo to create a directory under /opt/git/project.git. Change the owner to 'git'. Cd to the directory and initialize it.
  3. Create two virtual machines (vb1 and vb2). Creating a username 'david' on vb1 and a user name 'joseph' on vb2.
  4. Create ssh key for both 'david' and 'joseph'. Ssh to copy their ssh keys to git account.
  5. Create a new directory on vb1 and initialize it. Run git commands to commit & push files to the server (no password is needed). Note that when we use 'git commit', git will ask to create a username and email by first running 'git config' command.
  6. Switch to vb2 and run git clone (no password is needed). The user can modify the code and commit & push files to the server.
  7. Run 'git log' to check if each user's name/email are shown on the log.

Create a git server (github like w/ web interface)

If we like to create a github-like web interface, check out GitLab.

Below is my note

  1. https://about.gitlab.com/downloads/ contains steps of setting up Gitlab.
  2. By default, the domain name you have entered in setting up gitlab will be the URL you will use to access gitlab.
  3. Use the recommended method to install gitlab. Nginx will be installed as an http server.
  4. The root username and password is root and 5iveL!fe.
  5. When new users are created by root, we can put a faked email there (eg [email protected]). The root account can create password for the user.
  6. User's password is used to access GitLab web interface only. It is not used for pushing commits.
  7. After a new user is created, log out of root account and log in using the new user account. Click 'Profile setting' icon and then select SSH > Add SSH key. Copy your <id_rsa.pub> content there. To create your ssh key, use the command line "ssh-keygen -t rsa". The <id_rsa.pub> is located under ~/.ssh directory. The title should be auto populated. If ssh key is added successfully to gitLab, we won't get a pop-up asking password when we run 'git push'.
  8. A new project should be created by users (not root). If I create a project by root, I keep getting a permission issue when I run 'git push'.
  9. The username will affect path to all personal projects; e.g. [email protected]:newuser/test2.git.

Gitlab2.png Gitlab1.png Gitlab3.png

gitk - graphical tool to display history

General resources

Bitbucket (free for 5 users)

Bitbucket 101

  • Unlimited private repos
  • Code reviews
  • JIRA integration
  • REST API
  • Custom domains

See this post to know how to fix the problem of unknown author. In short, when I uncheck "Use global user setting" from Repository-> Repository Settings -> Advanced does the commit author change as expected.

See here for a list of Android apps related to bitbucket.

Convert an image file to a different format (eg icon)

Using imagemagic program. 

convert winamp-ncrow.png -resize 32x32 winamp-ncrow.ico

login shell vs interactive shell

8 http://serverfault.com/questions/8882/what-is-the-difference-between-a-login-and-an-interactive-bash-shell

  1. login shell - non desktop environment. ~/.bash_profile is sourced for the bash shell.
  2. interactive shell - Ctrl+Alt+t to open a terminal from a graphical mode (desktop environment). ~/.bashrc is source. We usually edit ~/.bashrc to set up the environment to include fancy prompt, set aliases, set history options, or define custom shell functions.

export environment variables

  1. Both a login shell and an interactive one. SSH (Putty) to connect to a remote machine.
  2. When a shell runs a script or a command passed on its command line, it's a non-interactive, non-login shell.

History of commands

history command with date and time

Running the following code once and history will give date and time the next time you issue the history command. 

echo 'export HISTTIMEFORMAT="%d/%m/%y %T "' >> ~/.bashrc

Note that the original post asks to write the line to ~/.bash_profile but this is not working in the desktop environment.

Bang bang - Run a command/Fetch parameters from previous history

  1. ^P: Move up through the command history list one command at a time.
  2. ^N: Move down through the command history list one command at a time.
  3. !!: Run the previous command.
  4. !n: Run command number n
  5. !string: Run most recent command starting with characters in string
  6. !?string: Run most recent command containing characters that match string
  7. !!*: Fetch parameters from last command

For example, 

!-1
!4

Listen to HiChannel internet radio

Use Radio Tray

I use it to listen m3u file (VLC also supports it too).

Web Analytics Reporting Tools

Paint.NET like program

Pinta

Calibre - Read ebook in epub format

$ sudo -v && wget -nv -O- https://github.com/kovidgoyal/calibre/raw/master/setup/linux-installer.py | sudo python -c "import sys; main=lambda:sys.stderr.write('Download failed\n'); exec(sys.stdin.read()); main()"
2014-03-19 15:54:28 URL:https://raw.github.com/kovidgoyal/calibre/master/setup/linux-installer.py [25423/25423] -> "-" [1]
Installing to /opt/calibre
Downloading tarball signature securely...
Will download and install calibre-1.28.0-x86_64.tar.bz2 
                                Downloading calibre-1.28.0-x86_64.tar.bz2                                
100% [===============================================================================================]
                                                                                                         Downloaded 63255897 bytes 
Checking downloaded file integrity... 
Extracting files to /opt/calibre ...
Extracting application files... 
Creating symlinks...
	Symlinking /opt/calibre/fetch-ebook-metadata to /usr/bin/fetch-ebook-metadata
	Symlinking /opt/calibre/lrf2lrs to /usr/bin/lrf2lrs
	Symlinking /opt/calibre/ebook-convert to /usr/bin/ebook-convert
	Symlinking /opt/calibre/ebook-meta to /usr/bin/ebook-meta
	Symlinking /opt/calibre/ebook-edit to /usr/bin/ebook-edit
	Symlinking /opt/calibre/lrfviewer to /usr/bin/lrfviewer
	Symlinking /opt/calibre/calibre to /usr/bin/calibre
	Symlinking /opt/calibre/markdown-calibre to /usr/bin/markdown-calibre
	Symlinking /opt/calibre/calibre-debug to /usr/bin/calibre-debug
	Symlinking /opt/calibre/calibre-parallel to /usr/bin/calibre-parallel
	Symlinking /opt/calibre/web2disk to /usr/bin/web2disk
	Symlinking /opt/calibre/calibre-server to /usr/bin/calibre-server
	Symlinking /opt/calibre/calibre-customize to /usr/bin/calibre-customize
	Symlinking /opt/calibre/ebook-polish to /usr/bin/ebook-polish
	Symlinking /opt/calibre/ebook-viewer to /usr/bin/ebook-viewer
	Symlinking /opt/calibre/calibre-smtp to /usr/bin/calibre-smtp
	Symlinking /opt/calibre/lrs2lrf to /usr/bin/lrs2lrf
	Symlinking /opt/calibre/ebook-device to /usr/bin/ebook-device
	Symlinking /opt/calibre/calibredb to /usr/bin/calibredb
Setting up command-line completion...
Installing bash completion to /etc/bash_completion.d/calibre
Setting up desktop integration...
Creating un-installer: /usr/bin/calibre-uninstall
Run "calibre" to start calibre

Fetch News

Several built-in sources like TIME, US NEWS report do not work. It'll show 'http error 404 not found'. The self added RSS feed method (see below) does work!

export DISPLAY

If we want to run a GUI app on a remote computer (such as Raspberry Pi/Beaglebone Black) using ssh, we can issue the following command before running the app. 

export DISPLAY=:0.0

LVM Demystified

http://www.linuxjournal.com/content/lvm-demystified

Shared library management

http://www.cyberciti.biz/tips/linux-shared-library-management.html

  1. ldconfig : Updates the necessary links for the run time link bindings.
  2. ldd : Tells what libraries a given program needs to run.
  3. ltrace : A library call tracer.
  4. ld.so/ld-linux.so: Dynamic linker/loader.

Install binary software using sudo

One example (Calibre) is like 

sudo -v && wget -nv -O- https://raw.githubusercontent.com/kovidgoyal/calibre/master/setup/linux-installer.py | \
sudo python -c "import sys; main=lambda:sys.stderr.write('Download failed\n'); exec(sys.stdin.read()); main()"

Note that in wget the option "-O-" means writing to standard output (so the file from the URL is NOT written to the disk) and "-nv" means no verbose.

If the option "-O-" is not used, we'd better to use "-N" option in wget to overwrite an existing file.

See the Logging and Download options in wget's manual. 

       -O file
       --output-document=file
           The documents will not be written to the appropriate files, but all
           will be concatenated together and written to file.  If - is used as
           file, documents will be printed to standard output, disabling link
           conversion.  (Use ./- to print to a file literally named -.)

Log files

DigitalOcean 

ls -lt /var/log

Linux command similar to top to show hard disk activity

Use iotop. On ubuntu, we can use sudo apt-get install to install it. Use sudo iotop to launch it. 

sudo iotop -o -u $USER

Install Apache HBase

Follow the Quick Start to downloaded hbase tar ball. Suppose we save the tar ball under ~/Downloads folder and extract it in the same directory. We shall edit conf/hbase-site.xml file according to their instruction. The following is my case. 

$ tar xzvf hbase-0.98.3-hadoop2-bin.tar.gz 
$ cd hbase-0.98.3-hadoop2/
$ cat conf/hbase-site.xml 
<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<configuration>
  <property>
    <name>hbase.rootdir</name>
    <value>file:///home/brb/Downloads/hbase-0.98.3-hadoop2/hbase</value>
  </property>
  <property>
    <name>hbase.zookeeper.property.dataDir</name>
    <value>/home/brb/Downloads/hbase-0.98.3-hadoop2/zookeeper</value>
  </property>
</configuration>

Before we follow the getting started guide to launch HBase, we shall make sure JAVA_HOME environment variable is created. 

$ ls /usr/lib/java
$ export JAVA_HOME=/usr/lib/jvm/java-6-openjdk-amd64

Note that the last line may be replaced by 

export JAVA_HOME=$(readlink -f /usr/bin/javac | sed "s:bin/javac::")

Then we can launch HBase, 

$ ./bin/start-hbase.sh starting master, logging to /home/brb/Downloads/hbase-0.98.3-hadoop2/bin/../logs/hbase-brb-master-brb-P45T-A.out
brb@brb-P45T-A:~/Downloads/hbase-0.98.3-hadoop2$ ./bin/hbase shell
2014-07-06 09:51:34,621 INFO  [main] Configuration.deprecation: hadoop.native.lib is deprecated. Instead, use io.native.lib.available
HBase Shell; enter 'help<RETURN>' for list of supported commands.
Type "exit<RETURN>" to leave the HBase Shell
Version 0.98.3-hadoop2, rd5e65a9144e315bb0a964e7730871af32f5018d5, Sat May 31 19:56:09 PDT 2014

hbase(main):001:0> create 'test', 'cf'
2014-07-06 09:51:49,510 WARN  [main] util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
0 row(s) in 2.0770 seconds

=> Hbase::Table - test
hbase(main):002:0> list 'test'
TABLE                                                                           
test                                                                            
1 row(s) in 0.0530 seconds

=> ["test"]
hbase(main):003:0> exit

brb@brb-P45T-A:~/Downloads/hbase-0.98.3-hadoop2$ ./bin/hbase shell2014-07-06 09:53:37,480 INFO  [main] Configuration.deprecation: hadoop.native.lib is deprecated. Instead, use io.native.lib.available
HBase Shell; enter 'help<RETURN>' for list of supported commands.
Type "exit<RETURN>" to leave the HBase Shell
Version 0.98.3-hadoop2, rd5e65a9144e315bb0a964e7730871af32f5018d5, Sat May 31 19:56:09 PDT 2014

hbase(main):001:0> list 'test'
TABLE                                                                           
2014-07-06 09:53:44,373 WARN  [main] util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
test                                                                            
1 row(s) in 1.4800 seconds

=> ["test"]
hbase(main):002:0> put 'test', 'row1', 'cf:a', 'value1'
0 row(s) in 0.4460 seconds

hbase(main):003:0> put 'test', 'row2', 'cf:b', 'value2'
0 row(s) in 0.0140 seconds

hbase(main):004:0> put 'test', 'row3', 'cf:c', 'value3'
0 row(s) in 0.0050 seconds

hbase(main):005:0> scan 'test'
ROW                   COLUMN+CELL                                               
 row1                 column=cf:a, timestamp=1404654837532, value=value1        
 row2                 column=cf:b, timestamp=1404654856976, value=value2        
 row3                 column=cf:c, timestamp=1404654866298, value=value3        
3 row(s) in 0.0560 seconds

hbase(main):006:0> get 'test', 'row1'
COLUMN                CELL                                                      
 cf:a                 timestamp=1404654837532, value=value1                     
1 row(s) in 0.0280 seconds

hbase(main):007:0> disable 'test'
0 row(s) in 1.6050 seconds

hbase(main):008:0> drop 'test'
0 row(s) in 0.2290 seconds

hbase(main):009:0> exit
brb@brb-P45T-A:~/Downloads/hbase-0.98.3-hadoop2$

curl vs wget

sudo apt-get install curl

For example, the Download link at the National Geographic Travel Photo Contest 2014 works for curl but not wget. I can use curl with -o option but not wget with -o option in such a case.

wget to download a website

To download a copy of a complete web site, up to five levels deep ('-r' option). All files linked to in the documents are are downloaded to enable complete offline viewing ('-p' and '--convert-links' options). Instead of having the progress messages displayed on the standard output, you can save it to a log file with the -o option. 

wget -p --convert-links -r linux.about.com -o logfile

Apply a patch to source code

For example Tophat 2.0.12 compatibility with Samtools 1.0, 

brb@brb-VirtualBox:~/Downloads$ ls
support_for_tophat_1.patch  tophat-2.0.12  tophat-2.0.12.tar.gz

brb@brb-VirtualBox:~/Downloads$ grep -r -i "check_samtools" tophat-2.0.12/
tophat-2.0.12/src/tophat.py:def check_samtools():
tophat-2.0.12/src/tophat.py:        check_samtools()

brb@brb-VirtualBox:~/Downloads$ cp support_for_tophat_1.patch tophat-2.0.12/src/
brb@brb-VirtualBox:~/Downloads$ cd tophat-2.0.12/src/
brb@brb-VirtualBox:~/Downloads/tophat-2.0.12/src$ patch tophat.py < support_for_tophat_1.patch 
patching file tophat.py
Hunk #1 succeeded at 1540 (offset 3 lines).
Hunk #2 succeeded at 1563 (offset 3 lines).
brb@brb-VirtualBox:~/Downloads/tophat-2.0.12/src$ ls

Get external IP address

It seems there is no way to get the external IP address without not using external services. 

curl http://ipecho.net/plain; echo
# OR
curl ipv4.ipogre.com

Domain

  • WHOIS LOOKUP. For example, consider "r-pkg.org" domain,
    • The 'Sponsoring Registrar' shows who is the sponsoring registrar (eg GoDaddy.com).
    • The 'Registrant Name' shows who registered this domain.
  • IP-Lookup. For example, consider "r-pkg.org" domain.
    • The linux command line tool 'ping' shows the ip address.
    • When we use the ip address to search in the IP-Lookup, the IP owner info > 'Organization' or 'OrgName' field shows the owner of this IP (eg Digital Ocean, Inc).
    • The Domain owner info there gives the same (or less) information as WHOIS LOOKUP.
  • WhoIsHostingThis (eg Verizon, Amazon, GoDaddy, CloudFlare, Github, etc) and its wikipedia page.

Subnet

IPv4

  • Class A: 255.0.0.0 or /8;
  • Class B, 255.255.0.0 or /16;
  • Class C, 255.255.255.0 or /24.

For example, in the subnet 192.168.5.0/255.255.255.0 (192.168.5.0/24) the identifier 192.168.5.0 commonly is used to refer to the entire subnet.

In the /16 subnet 192.168.0.0/255.255.0.0, which is equivalent to the address range 192.168.0.0–192.168.255.255

Build a home network

Monitor harddisk health by command line

Using smartctl.

Troubleshoot and repair network problems

http://www.linuxuser.co.uk/features/troubleshot-repair-linux-networks

Monitor network by command line

Use iftop command. On ubuntu, we need to use sudo apt-get install iftop and then run it by sudo iftop -i eth0 -n where -n is to disable dns lookup. After that, we can press some keys to toggle options.

  • p: port
  • s: source
  • d: destination

Copy text to a clipboard to be used in other apps

Install the xclip program. See here or here. 

sudo apt-get install xclip
# Example
sort -n -k 3, -k 2 file.txt | xclip -selection clipboard

Works.

Start Emacs without X

Add -nw (no window) option. 

emacs -nw

Reduce the size of an mp3 file

Specify a new lower bitrate using the -b option in lame. For example if your starting mp3 has a quality of 256kbs you can lower its bitrate to 128kbps (or even lower like 64kbps) by: 

lame --mp3input -b 128 input.mp3 output.mp3

Normalize the volume of an audio file

  • Use Audacity. To raise (Amplify) volume: 1. Edit > Select All. 2. Effect > Amplify. Increase db from 0 to 15, for example. Check clip3. Export > MP3 or just start to listen.
  • Command line tool: avconv (replace ffmpeg program). See this post.
avconv -ss 00:00:10 -i OLD.mp3 -vol 2560 NEW.mp3

The anconv/ffmpeg -vol parameter amplifies the sound. The default value is 256 (no amplification), and you can adjust the number accordingly. Here it’s 2560, as it’s 10 times louder. Note that these are not decibel values or anything that sophisticated, but just an integer value. 512 equals to twice the volume, 768 three times, 1024 four times, etc. The -ss parameter specifies the start time offset. Here it will skip the first 10 seconds.

I found the converted file by sox has about one half file size compared to anconv/ffmpeg program (source file=47MB, anconv converted=135MB, sox converted file=54MB). 

sudo apt-get install sox libsox-fmt-all
sox --norm OLD.mp3 NEW.mp3 trim 10
sox --norm OLD.mp3 NEW.mp3 silence 1 0.1 1%
sox -v 4.0 OLD.mp3 NEW.mp3             # increase volume

where '--norm' will normalize the audio and the 'trim' option set to skip the first 10 seconds. The silence parameter allows to trim silence at the beginning without a need to specify the number of seconds.

Wireshark

Track the Time a Command Takes

Use time command (See `man time`). 

time COMMAND
time (COMMAND1; COMMAND2)
time (COMMAND1 && COMMAND2)

Magazines

PDF reader

The default one Evince seems slow when I try to view odroid magazine. I installed and tried MuPDF (github source code). It seems faster and I don't see blank pages when I view one odroid magazine. In terms of speed, mupdf >> xpdf >> okular >> Evince. 

sudo apt-get install mupdf

Other choices are

  • acroread
  • xpdf
  • okular (KDE/Qt application), allow to change its background color
  • kpdf
  • gv
  • qpdfview
  • Foxit or PDF-XChange Viewer(needs wine)

PDF highlight and annotation

Install Okular by 

sudo apt-get install okular

To highlight a line, click F6 (Tools -> Review) to turn on the annotation tool bar (it will be shown on the left hand side of the documentation). You can then click

  1. the 4th icon to highlight a line (it may not be able to select the right texts we want. But when it works the result is nice)
  2. the last icon to draw an ellipse or a rectangle (to change from an ellipse to a rectange you can click Settings -> configure Okular... -> annotation)

Another method is to use a windows program and run it using Wine. See the discussion here.

dclock

Digital clock for the X Window System with flexible display. 

sudo apt-get install dclock
dclock -h
dclock -d
dclock -date "Today is %A %B %Y" -geometry 577x194+119+139

wine and winetricks

Running Linux in the AWS/Amazon Web Services

Debian/Ubuntu/Bio-Linux

Debian

Download Debian

Go to http://www.debian.org/distrib/ and download "small installation image", for example debian-8.0.0-amd64-netinst.iso (release Apr 2015). It is about 246 MB.

Screenshots of step-by-step installation can be found in here.

At the end of installation, it will offer a collection of software to install. Below 'Debian desktop environment', it has a selection of GNOME, Xface, KDE, Cinnamon, MATE and LXDE (new in Jessie/8.0). Note that the cd images download page only provides a selection of kde, lxde and xfce desktop. The default is 'GNOME' in Jessie.

After installation, you got a desktop environment of Debian based on GNOME 3 (Virtual machine will use recovery mode, but still works. Some people suggest to install the guest additions (in the guest) and make sure that you enable the 3D acceleration in the guest settings.). Also 'free -m' command shows it uses 202 MB memory and the whole system takes up 3.3 GB. I am testing on a Chinese desktop environment.

When Debian is in recovery mode, the desktop interface is like old fashion. Application and Place on top of the screen. When GNOME 3 is working (in my test of Debian 7.1.0, I cannot run VBoxLinux.run, but it still works after I did other steps), the interface is sort of Ubuntu with application launched from the left hand side. It differs from Ubuntu because the side bar appears by clicking a 'preview' button on top left corner.

The default browser in Debian is Iceweasel with AdBlock Plus preinstalled.

I don't know why the default user does not have root privilege.

When I installed the Chinese version, the keyboard switch icon (SCIM) is automatically available. However once the desktop is in regular GNOME 3, the switch icon disappeared. Fortunately, we can use Ctrl + Space to switch languages. Thanks to the hint there.

Server version

There is no a server version of iso to download. At the end of installation, it will ask what software to install: Debian desktop environment, Web server, SSH server, Laptop, SQL database, ... We can uncheck 'Debian desktop environment' item.

Compared to Desktop version, the server version takes 1.3GB space and 33MB memory.

The server version does not have 'sudo' command. Use 'su' to switch to 'root' user.

Note that even we installed 'sudo', we can not use 'sudo' from the default user. It will complain xxx is not in the sudoers file. This incident will be reported..

Browse iso files

Note that if we want to download the iso image, we should consider using the torrent method. The can see a variety of download options from

http://www.debian.org/CD/ > Downloading Debian CD/DVD images via HTTP/FTP

Permission denied

http://roger.steneteg.org/blog/virtualbox-guest-additions-on-debian/

The script uses /bin/sh as shell and on Debian Wheezy/Jessie /bin/sh is symlinked to /bin/dash. Dash is a more light-weight replacement for Bash, and it turns out that the VirtualBox script does not work as it should when run with "dash".

An easy workaround is to explicitly run the script with "bash" with the following command: 

sudo bash ./VBoxLinuxAdditions.run

Ubuntu/Kubuntu/Lubuntu/Xubuntu

Download links for all versions from wiki.ubuntu.com > releases.ubuntu.com.

Ubuntu flavors and derivatives

For some reason, when I try the Ubuntu (13.04) live CD, the screen resolution looks perfect. But when I installed the OS, the screen resolution is always too low. The propriety graphics driver cannot be installed successfully. Fortunately, when I try the Kubuntu (13.04), the display resolution problem automatically works!

Update: Kubuntu failed to respond after I install SCIM related programs. A freshly installed linuxmint OS also has a similar problem that the desktop does not respond to mouse or keyboard. Luckily, the Xubuntu works fine and the Chinese input works out of box if I choose Chinese as desktop evvironment (339MB was used).

Black screen on boot Ubuntu 14.04

Press e when you see the first menu.

Server version

UbuntuServerInstall.png

AV Linux

AV Linux features a complete customized Debian Linux XFCE4 4.10 Desktop Environment with the added bonus of a handpicked selection of pre-tested and pre-configured Audio, Graphics and Video content creation software demonstrating the excellence of Open-Source and also includes many unique Commercial Demos.

Selection of desktop environment

  • Unity - Original Ubuntu. Use 'unity --version' to check the unity version. If something was screwed up (eg after we remove gnome-desktop), we can reinstall the unity desktop by
sudo apt-get install ubuntu-desktop
  • GNOME - Ubuntu GNOME (GNOME 3). The build-in screensaver is a digital clock showing the current time & date. Cool! This seems to be a new feature in GNOME 3.6 optimized for touch screen devices. See this and this.

UbuntuGnome1404.png UbuntuGnome1404b.png

Note that we can install the gnome desktop by using the command line. It will keep the current wallpaper. The clock in screensaver will not be shown until we shake the mouse or keyboard. 

sudo apt-get install gnome-shell
# Choose 'gdm' (Gnome Desktop Manager) as the display manager instead of 'lightdm' the Ubuntu's default
#    when it is configuring gdm as only GDM offers GNOME-specific features such as lock-screen notifications.
# See the screenshot at 
#    https://ideasnet.wordpress.com/2013/05/11/ides-desktop-how-to-replace-unity-with-gnome-3-8-in-ubuntu-13-04-desktop-edition/
# If messed up, run "sudo dpkg-reconfigure gdm"
sudo apt-get install ubuntu-gnome-desktop

If the lock screen does not work, use Settings > Brightness and Lock, or use the command line 

gsettings set org.gnome.desktop.lockdown disable-lock-screen 'false'

If the screensaver is not working, try 

sudo apt-get install gnome-screensaver

In my case, the screen turns off (black). But if we wake the PC up, the time and date screen shows up.

Some differences (inconvenience): 1. No maximize, minimize windows buttons 2. Have to click 'Activities' button (or 'Windows' key) to switch applications. These complains also appeared in other review.

  • KDE - Kubuntu
  • Xfce - Xubuntu. The response is quicker when I compare the speed by clicking the top-left icon (app menu) in Xfce and Acitvities button in ubuntu-GNOME. This is tested when both Xubuntu and ubuntu-GNOME are installed in VirtualBox.
  • LXDE - Lubuntu, LXLE and LXQt. LXDE is the default desktop environment for Raspbian, LXLE, BBB, and Lubuntu.
  • Cinnamon - Linux Mint. A GTK+-based desktop environment. Note that Mint releases only LTS versions (5 years support as Ubuntu).
  • MATE - Ubuntu Mate. The MATE Desktop Environment is the continuation of GNOME 2.
  • Pantheon - Elementary OS. I cannot make the Chinese input to work (I can install ibus-chewing but cannot switch input methods?).

To query the desktop environment using the command line, use 

echo $DESKTOP_SESSION

On Ubuntu, it returns 'ubuntu' (gnome 3+unity), on Debian 8.0 it shows 'default' and on Debian's BBB, it returns 'LXDE'.

Determine/install/switch Window Manager

sudo apt-get install wmctrl
wmctrl -m           

sudo apt-get install <pkg-name>
<pkg-name> --replace

On Ubuntu the default window manager is Compiz, for xubuntu it is Xfwm4 and for BBB it is Openbox.

After running update/upgrade in Ubuntu

  • Virtualbox does not work. After initial fix, the guest machine cannot connect to internet

Ubuntu Software Repository

The repository components are:

  1. Main - Officially supported software.
  2. Restricted - Supported software that is not available under a completely free license.
  3. Universe - Community maintained software, i.e. not officially supported but enabled by default software.
  4. Multiverse - Software that is not free.

See the pitfall in the PCWorld article.

"Failed to download Package Files" - Software Updater

One solution is to run the following command first 

sudo apt-get update && sudo apt-get upgrade

and then run the software updater. Usually it requires the computer to restart.

I personally adjust the frequency of notification by choosing 'Every two weeks' for Automatically check for updates, etc.

apt-get upgrade vs apt-get dist-upgrade

My experience aligns with the statement: dist-upgrade is more likely to break stuff badly than upgrade..

Customize the desktop

  • Install Cairo-Dock.

Create customized ubuntu iso

Grub2

To show the grub2 screen, run 'sudo nano /etc/default/grub' and comment out the line GRUB_HIDDEN_TIMEOUT=0 and change the line GRUB_CMDLINE_LINUX_DEFAULT="quiet splash" to GRUB_CMDLINE_LINUX_DEFAULT="". Remember to run 'sudo update-grub' after any change to grub.

To add a splash image, follow the instruction at https://help.ubuntu.com/community/Grub2/Displays. Note that Grub2 will search the image based on some priority and there are also some minor requirements on the images. To test 

sudo apt-get install grub2-splashimages
sudo cp /usr/share/images/grub/Moraine_Lake_17092005.tga /boot/grub/
sudo update-grub

Boot from an NSF server

With NFS booting, the core kernel and file systems are kept on a central server and then pushed out onto client systems to be booted on there. That means your files and desktop will always be available wherever you want to log in.

At the end, the client computer does not need any internal storage. Cool!

Change the default session when using auto login

See this post. We need to edit the file /etc/lightdm/lightdm.conf. Note that on my Ubuntu 12.04, I have the following desktop options. 

$ ls -lt /usr/share/xsessions/
total 16
-rw-r--r-- 1 root root 205 Apr 19  2012 gnome.desktop
-rw-r--r-- 1 root root 188 Apr 19  2012 gnome-shell.desktop
-rw-r--r-- 1 root root 208 Apr 19  2012 ubuntu-2d.desktop
-rw-r--r-- 1 root root 185 Apr 19  2012 ubuntu.desktop

Debian root user from remote access

When you login by SSH, then use the username you have chosen when you installed Debian as the root user is disabled for remote logins. Then run the command "su" to become root user. See howtoforget.com.

Terminal tab color

If we open several tabs in the (GNOME) Terminal, the default color of the active tab is not quite different from the other tabs. We need a solution to change the tab colors.

http://askubuntu.com/questions/355297/gnome-terminal-tabs-no-contrast-between-active-and-inactive-tabs 

Solution: edit ~/.config/gtk-3.0/gtk.css (you might have to create it) and add:

TerminalWindow,
TerminalWindow.background {
        background-color: #6e6e6e;
        color: #000000;
}

TerminalWindow .notebook tab {
        padding: 2;
        background-color: #6e6e6e;
}

TerminalWindow .notebook tab:active {
        background-color: #d1d1d1;
}

Then close ALL terminal windows start and test.

Search package full name

Use apt-file command. See this post

Find package information before installing it

Use "-s" option for simulation. No sudo is necessary. 

apt-get -s install PACKAGENAME

The output is too much (include other dependences)

A better way is to use aptitude (which is not installed by default in Ubuntu) 

aptitude search <package> -F "%c %p %d %V"

For example, 

debian@beaglebone:~/qt-4.8.5/bin$ aptitude search  qtcreator -F "%c %p %d %V"
p qtcreator                         lightweight integrated development environme 2.5.0-2       
p qtcreator:armel                   lightweight integrated development environme 2.5.0-2       
p qtcreator-dbg                     debugging symbols for Qt Creator IDE         2.5.0-2       
p qtcreator-dbg:armel               debugging symbols for Qt Creator IDE         2.5.0-2       
p qtcreator-doc                     documentation for Qt Creator IDE             2.5.0-2

If a package is already installed, we can use the following way to check version number. 

apt-show-versions <package>

Install Microsoft Font in linux suite

PCWorld.com

Change locale language and character set

Display Chinese character (from vanilla Debian/Ubuntu system)

sudo apt-get install fonts-arphic-ukai fonts-arphic-uming

sudo apt-get install language-support-fonts-zh
# Or
sudo apt-get install ttf-arphic-uming ttf-wqy-zenhei

Chinese Input

http://pinyinjoe.com/linux/ubuntu-12-chinese-setup.htm

Note: If ibus does not have chewing selection, we should install it first. 

sudo apt-get install ibus-chewing
  1. type 'language' in Dash search (Or click Power button on the top-right corner -> System Settings...) and click 'Language Support' (this may not be installed from (x)Ubuntu automatically. In such case, IBUS will be missing eg Chewing method). The 'Language Support' is under Settings menu in xUbuntu.
  2. choose 'IBUS' for keyboard input method. IBUS is under 'System' menu in xUbuntu.
  3. Click 'Install/Remove Languages'. Choose Chinese -> Chewing. Note that Chewing is not the same as Bopomofo. I cannot get used to Bopomofo method.
  4. Settings > Keyboard Input Method > Input method > Select an input method > Show only input methods for your region > Chinese 酷 Chewing. If we cannot find this dialog, we can launch it by ibus-setup command. If Chewing is not shown as one of Chinese input methods, we can log out and log in the desktop.
  5. Click 'Close' button.

Log out and Log in. Click Keyboard icon on top right corner. Modify its Preferences. Go to 'Input Method' tab. Add Chinese-Chewing and click 'Close' button. We can use Ctrl+Space to switch language input.

On Ubuntu 13.10, type 'language support' in the Dash. Click 'Install/Remove Languages' and check Chinese. In the 'Keyboard input method system', change from default to 'IBus'. We shall see a keyboard icon on the top right of menu. Click it and choose Text Entry Settings. Add Chinese Chewing input. Then change 'switch to next source using' method to 'Ctrl + Space' by pressing both keys on your keyboard. Bingo!

Note that do not use 'Text Entry' application (shown as one options when we search 'language'). This application will conflict with ibus.

If for some reason, ibus-daemon was not started automatically, try the approach here by adding ibus-daemon to the startup application list.

Samba

sudo apt-get install samba samba-common
sudo apt-get install python-glade2
sudo apt-get install system-config-samba

Use Dash and search for 'samba'. It will ask for the user's password first. The samba password can also be set by 

sudo smbpasswd -a USERNAME

A non-gui way to configuration samba is adding the following to the end of /etc/samba/smb.conf file, sudo nano -w /etc/samba/smb.conf 

[brb]
        path = /home/brb
;       writeable = no
;       browseable = yes
        guest ok = yes

smb://192.168.1.3/

On Windows PC, go to start and open 'Run' then enter ip with double backslash. Like this (\\192.168.1.3).

Change time zone

http://wiki.debian.org/TimeZoneChanges 

$sudo dpkg-reconfigure tzdata

The command launches an ncurses based interface which allows arrow keys to be used to select the region specific time zone.

Make script run at boot time with init.d directory

http://www.debian-administration.org/articles/28

For example, see here from running a python script for raspberry pi.

  1. Create a script /etc/init.d/lcd
  2. Make the script executable
  3. Make the script known to the system by using the update-rc.d' command
sudo update-rc.d lcd defaults

wireless connection randomly drop off

My wireless adapter is TP-LINK, TL-WN722N.

http://askubuntu.com/questions/73607/wireless-connection-keeps-dropping-with-an-intel-3945abg-card 

sudo iwconfig wlan0 power off

5 Best Linux/Ubuntu compatible USB Wifi cards:

  • AirLink101 AWLL6075 Wireless N Mini USB Adapter
  • Medialink – Wireless N USB Adapter – 802.11n
  • ASUS (USB-N13) Wireless-N USB Adapter
  • Panda Mini Wifi (b/g/n) 150Mbps Card
  • TP-Link TL-WN722N 150Mbps High Gain Wireless USB Adapter

To show (USB) wireless adapter information 

sudo lshw -C network
sudo lsusb -v
iwconfig

To check wifi connection information (SSID, channel, address, frequency, qualiyt, signal level, ...) 

sudo iwlist wlan0 scan

My experience is quality should be at least 50/70 and signal level should be larger than -60dBm.

Write an iso or img file to a USB flash drive

Official instruction on www.ubuntu.com

Use dd

First, get to know the USB drive device name like /dev/sdb. When using 'dd', the USB drive has to be unmounted (using 'umount' command, not click 'reject' button in File Manager). Note that this will irrevocably destroy all data on /dev/sdX.

The instruction can be found in a lot of places like Archlinux wiki page. 

sudo fdisk -l
sudo dd bs=4M if=xxx.img of=/dev/sdb && sync

where /dev/sdb is a device name, not a partition name. We can also adjust bs to a smaller value like 1M, 4m.

For some reason when I use dd to create ubuntu 14.04 on usb drive, sudo gparted also gives me a Libparted warning /dev/sdc contains GPT signature, indicating that it has a GPT table. However, it does not have a valid fake msdos partition table, as it should... Is it a GPT partition table? messsage. If I click 'Yes', Gparted shows no partition on the usb drive??? Nevertheless, the usb drive can be used to boot into ubuntu 14.04.

UNETBOOTIN

The GUI is written by Qt so the program is cross-platform. See its wiki.

http://xubuntugeek.blogspot.com/2013/04/how-to-install-xubuntu-on-usb-device.html 

sudo apt-get install unetbootin extlinux -y

The following is a screenshot of the contents of xubuntu 12.04. The usb drive needs to be formatted to fat32 on Windows OS to repair partition table error. The partition table error was discovered when I use sudo gparted program to view the USB drive.

Note that Unetbootin (Windows & Linux) and Universal USB installer (Windows only) are quite similar although Universal USB installer provides more options in its interface while Unetbootin does not have any other options.

XUbuntu12044 USBdrive.png XUbuntu gparted.png

Universal USB Installer/UUI

http://www.pendrivelinux.com/universal-usb-installer-easy-as-1-2-3/

I first used dd command to successfully created a xubuntu 14.04 usb drive but there seems to be a problem with the partition since the gparted program will give a warning message about that and also the whole 16GB was used when I insert the drive onto a Windows machine.

Note that the fdisk utility cannot handle this new partition format so we have to use the gparted program.

Open the gparted program. Use it MULTIPLE TIMES to create a partition table (Device -> Create Partition Table...). Then we create a FAT32 partition with all of the space. After that, we can use the UUI program to create an Ubuntu USB drive.

The following screenshots are from a 16GB USB drive.

UUIa.png UUIb.png

Create an iso file from a CD or HD

Method 1 (Better than Method 2). No need to umount the cdrom. 

mkisofs -o /tmp/DVD.iso /tmp/directory/

to make an ISO from files on your hard drive.

Method 2. Make sure the cdrom is NOT mounted. Type mount to confirm it. If cd was mouted automatically unmout it with umount command: like umount /dev/cdrom or umount /mnt/cdrom. Note that no extra forward slash after /dev/cdrom for the command below. 

dd if=/dev/cdrom of=~/DVD.iso

We can mount the iso file to a directory to check the iso file content is correct. 

mount -t iso9660 -o loop,ro DVD.iso /mnt

Have fun with /etc/hosts file

su -c "nano /etc/hosts" 

74.125.67.100   DNS_NAME1 DNS_NAME2

Find a file based on file name

$ find . -iname '*.txt'

Find files modified in one day. 

$ find . -mtime -1

Find files modified in one day and contain string 'est' 

$ find . -mtime -1 -exec grep --with-filename est {} \;

If the search directory is not the current directory, we need to add a forward slash to the directory name. 

$ find ~/Desktop -iname '*.txt'  # Not working
$ find ~/Desktop/ -iname '*.txt' # Working

The following example shows we can list multiple search criteria. The “‑r” option in tar appends files to an archive. xargs is a handy utility that coverts a stream of input (in this case the output of find) into command line arguments for the supplied command (in this case tar, used to create a backup archive). 

find / -type f -mtime -7 | xargs tar -rf weekly_incremental.tar
gzip weekly_incremental.tar

Find a file by searching contents

grep -r -i "Entering" ~/Downloads/R-3.0.0/

where -r means recursively searching the directory and -i means case insensitive.

A GUI version of a tool to search files is searchmoney. On ubuntu, we install it by 

sudo apt-get install searchmonkey

https://www.howtoforge.com/tutorial/linux-grep-command/ gives more examples

  • Using grep to search only for words ("-w" option)
  • Using grep to search two different words (egrep -w 'word1|word2' /path/to/file)
  • Count line for matched words ("-c" option)
  • Grep invert match ("-v" option)
  • How to list only the names of matching files ("-l" option)

Mount a remote file system over ssh

The trick is to use sshfs tool.

On Ubuntu 

# Install the program
sudo apt-get install sshfs

# Mount the file system
sudo mkdir /mnt/droplet <--replace "droplet" whatever you prefer
sudo sshfs [email protected]:/ /mnt/droplet

# Unmount the file system
sudo umount /mnt/droplet

# Permanently Mounting the Remote File System
sudo nano /etc/fstab
sshfs#[email protected]:/ /mnt/droplet

Nautilus (File Manager)

Undo Ctrl+L

Press ESC.

Mount another Linux system in Nautilus

Very easy. Check out howtogeek.com

Create a desktop shortcut

Navigate to your application in Nautilus. Right-click, select "Make Link". Then drag shortcut to your desktop. Works in Ubuntu 12.04.

Mount iso file

$ sudo mkdir /mnt/iso
$ sudo mount -o loop /tmp/file.iso /mnt/iso
$ sudo umount /mnt/iso

Check ubuntu version from command line

http://www.howtogeek.com/206240/how-to-tell-what-distro-and-version-of-linux-you-are-running/ 

lsb_release -a
# OR
cat /etc/issue
# OR
cat /etc/*release
# check kernel version
uname -r
# check 32/64 bit kernel
uname -a

keyboard shortcuts

Go to keyboard app to change the settings. Note: Super key is also Windows key. Use 'Backspace' key to disable a shortcut.

  • List from ubuntu.com
  • Super: Open the Dash. Press and hold it to see a cheat sheet with a bunch of other nifty shortcuts.
  • Super + Number: Open the application that is at that position in the dock
  • Alt + F2: Run an application by typing its name in the box which appears (same function as the Super key).
  • Alt + Space: Activate the window menu. Not useful so I disable it.
  • Ctrl + Super + D: To minimize all windows
  • Ctrl + Alt + Arrows: move to another workspace
  • Ctrl + Alt + Shift + Arrows: move current application to another workspace

On Xubuntu,

  • Alt + F1 (Ctrl + ESC): Application menu.
  • Alt + F2 (Super + R): Application Finder.

BioLinux

Bio-Linux 7.0 (2012/11/21) screenshot.

Biolinux can be installed in two ways.

  1. One is to download iso image file. http://nebc.nerc.ac.uk/downloads/
  2. The other way is to install Bio-linux software/package by using apt-get install method. See http://nebc.nerc.ac.uk/tools/bio-linux/other-bl-docs/package-repository

BioLinux.png

Check Biolinux version

cat /etc/bio-linux.version  # 8.0.5 as of June 2015

Installation

wget -qO- http://nebc.nerc.ac.uk/downloads/bl8_only/upgrade8.sh | sudo sh

This takes a long long time.

Software list

FAQ

Sample Data

Did not find them useful. 

brb@biolinux[brb] ls Desktop/Sample\ Data/                                                         [10:05AM]
act       cytoscape               glimmer3      mrbayes             peptide_seqs       splitstree
artemis   dendroscope             happy         mspcrunch           phylip             squint
blast     dotter                  hmmer         multiple_alignment  qiime              t-coffee
blast+    dust                    jalview       mummer              rasmol             tree-puzzle
blixem    fasta                   jprofilegrid  muscle              rdp_classifier     treeview
cap3      fastDNAml               mesquite      mview               readseq
catchall  forester-archaeopteryx  mira          njplot              samtools
clustal   gap4                    mothur        nucleotide_seqs     sanger_tracefiles
brb@biolinux[brb] ls Desktop/Sample\ Data/fasta/                                                   [10:05AM]
bovgh.seq        hahu.aa       mgstm1.e05           mgstm1.nt1r  myosin_bp.aa  oohu.raa
bovprl.seq       hsgstm1b.gcg  mgstm1.eeq           mgstm1.nts   n0.aa         prio_atepa.aa
egmsmg.aa        hsgstm1b.seq  mgstm1.esq           mgstm1.raa   n1.aa         prot_test.lib
grou_drome.pseg  humgstd.seq   mgstm1.gcg           mgstm1.rev   n2.aa         prot_test.lseg
gst.nlib         lcbo.aa       mgstm1_genclone.seq  mgstm1.seq   n2_fs.lib     qrhuld.aa
gst.seq          m1r.aa        mgstm1.lc            mgtt2_x.seq  n2s.aa        sql
gstt1_drome.aa   m2.aa         mgstm1.nt            ms1.aa       n2t.aa        titin_hum.aa
gstt1_pssm.asn1  mchu.aa       mgstm1.nt1           mu.lib       n_fs.lib      titin_hum.seq
gtm1_human.aa    mgstm1.3nt    mgstm1.nt12r         musplfm.aa   ngt.aa        xurt8c.aa
gtt1_drome.aa    mgstm1.aa     mgstm1.nt13          mwkw.aa      ngts.aa       xurt8c.lc
h10_human.aa     mgstm1.aaa    mgstm1.nt13r         mwrtc1.aa    oohu.aa       xurtg.aa

CloudBioLinux

Apache configuration

https://help.ubuntu.com/11.10/serverguide/httpd.html

Device manager

By default, ubuntu does not provide any graphical tool like device manager on Windows. A very close one is lshw (hardware lister). A GUI tool based on it is called lshw-gtk and can be installed by sudo apt-get install lshw-gtk in Ubuntu/Debian or yum install lshw in Red Hat/Fedora/CentOS. 

$ sudo lshw -short
H/W path        Device      Class       Description
===================================================
                            system      To Be Filled By O.E.M. (To Be Filled By O.E.M.)
/0                          bus         970 Extreme4
/0/4                        processor   AMD Sempron(tm) 145 Processor
/0/4/5                      memory      128KiB L1 cache
/0/4/6                      memory      1MiB L2 cache
/0/10                       memory      8GiB System Memory
/0/10/0                     memory      2GiB DIMM DDR3 Synchronous 667 MHz (1.5 ns)
/0/10/1                     memory      2GiB DIMM DDR3 Synchronous 667 MHz (1.5 ns)
/0/10/2                     memory      2GiB DIMM DDR3 Synchronous 667 MHz (1.5 ns)
/0/10/3                     memory      2GiB DIMM DDR3 Synchronous 667 MHz (1.5 ns)
/0/0                        memory      64KiB BIOS
/0/100                      bridge      RD890 PCI to PCI bridge (external gfx0 port B)
/0/100/2                    bridge      RD890 PCI to PCI bridge (PCI express gpp port B)
/0/100/2/0                  display     RV770 [Radeon HD 4850]
/0/100/2/0.1                multimedia  RV770 HDMI Audio [Radeon HD 4850/4870]
/0/100/9                    bridge      RD890 PCI to PCI bridge (PCI express gpp port H)
/0/100/9/0                  bus         ASM1042 SuperSpeed USB Host Controller
/0/100/a                    bridge      RD890 PCI to PCI bridge (external gfx1 port A)
/0/100/a/0                  bus         ASM1042 SuperSpeed USB Host Controller
/0/100/11                   storage     SB7x0/SB8x0/SB9x0 SATA Controller [IDE mode]
/0/100/12                   bus         SB7x0/SB8x0/SB9x0 USB OHCI0 Controller
/0/100/12.2                 bus         SB7x0/SB8x0/SB9x0 USB EHCI Controller
/0/100/13                   bus         SB7x0/SB8x0/SB9x0 USB OHCI0 Controller
/0/100/13.2                 bus         SB7x0/SB8x0/SB9x0 USB EHCI Controller
/0/100/14                   bus         SBx00 SMBus Controller
/0/100/14.1                 storage     SB7x0/SB8x0/SB9x0 IDE Controller
/0/100/14.2                 multimedia  SBx00 Azalia (Intel HDA)
/0/100/14.3                 bridge      SB7x0/SB8x0/SB9x0 LPC host controller
/0/100/14.4                 bridge      SBx00 PCI to PCI Bridge
/0/100/14.5                 bus         SB7x0/SB8x0/SB9x0 USB OHCI2 Controller
/0/100/15                   bridge      SB700/SB800/SB900 PCI to PCI bridge (PCIE port 0)
/0/100/15.2                 bridge      SB900 PCI to PCI bridge (PCIE port 2)
/0/100/15.2/0               bus         VT6315 Series Firewire Controller
/0/100/15.3                 bridge      SB900 PCI to PCI bridge (PCIE port 3)
/0/100/15.3/0   eth0        network     RTL8111/8168B PCI Express Gigabit Ethernet controller
/0/100/16                   bus         SB7x0/SB8x0/SB9x0 USB OHCI0 Controller
/0/100/16.2                 bus         SB7x0/SB8x0/SB9x0 USB EHCI Controller
/0/101                      bridge      Family 10h Processor HyperTransport Configuration
/0/102                      bridge      Family 10h Processor Address Map
/0/103                      bridge      Family 10h Processor DRAM Controller
/0/104                      bridge      Family 10h Processor Miscellaneous Control
/0/105                      bridge      Family 10h Processor Link Control
/0/1            scsi2       storage     
/0/1/0.0.0      /dev/cdrom  disk        CDDVDW SH-S223Q
/0/2            scsi4       storage     
/0/2/0.0.0      /dev/sda    disk        80GB MAXTOR 4K080H4
/0/2/0.0.0/1    /dev/sda1   volume      66GiB EXT4 volume
/0/2/0.0.0/2    /dev/sda2   volume      8186MiB Extended partition
/0/2/0.0.0/2/5  /dev/sda5   volume      8186MiB Linux swap / Solaris partition
/0/3            scsi6       storage     
/0/3/0.0.0      /dev/sdb    disk        SCSI Disk
/0/3/0.0.1      /dev/sdc    disk        SCSI Disk
/0/3/0.0.2      /dev/sdd    disk        SCSI Disk
/0/3/0.0.3      /dev/sde    disk        SCSI Disk
/1              wlan0       network     Wireless interface

For storage part, ubuntu provides a graphical tool. See "disk utility" on gnome based ubuntu or search for "disk" in launcher.

Set static IP

root@debian:~# cat /etc/network/interfaces
auto lo eth0
iface lo inet loopback
iface eth0 inet dhcp

Now edit the file /etc/network/interfaces 

# The primary network interface
auto eth0
iface eth0 inet static
address 192.168.1.3
gateway 192.168.1.1
netmask 255.255.255.0
network 192.168.1.0                 # Optional
broadcast 192.168.1.255             # Optional
dns-nameservers 192.168.1.1 8.8.8.8 # Or skip 192.168.1.1

After it, restart the network by issuing 

/etc/init.d/networking restart

OR sudo reboot

Note: It does not work by editing /etc/resolv.conf since this file will be overwritten.

Change IP address from the command line

/sbin/ifconfig eth0 192.168.1.17 netmask 255.255.255.0 up
/sbin/ifconfig eth0

Dyndns

See https://help.ubuntu.com/community/DynamicDNS

nano /etc/ddclient.conf 

daemon=600
protocol=dyndns2
use=web, web=checkip.dyndns.com, web-skip='IP Address'
server=members.dyndns.org
login=YOURNAME
password='YOURPASSWORD'
taichi.selfip.net

And run ddclient -daemon=0 -debug -verbose -noquiet to confirm it is working.

Mail

  • check the option of Mail Routing: I have mail server with another name and would like to add MX hostname...
  • In 'MX hostname' entering aspmx.l.google.com
  • In 'Primary' choose 'Yes, use it as my primary mail relay.'

After that, see http://yaui.me/postfix-gmail-smtp-server-relay-ubuntu/ for setup on Ubuntu machine.

no-ip

Similar to Dyndns. It has its own client program. Needs to build it yourself.

Also see the troubleshooting guide.

See http://ducky-pond.com/posts/12 for instruction of setting autostart on Debian system.

See http://www.coulterfamily.org.uk/pages/PCs/Linux/FAQ-LINUX-NO-IP-CLIENT.php for another approach.

Note: If noip2 cannot start automatically or noip2 does not update even it can be seen from ps -ef command, use sudo crontab -e command. For some reason, after I use sudo crontab, noip2 can update IP. So the only problem right now is it cannot update every 30 minutes even sudo noip2 -S says so. The problems may be 1. ps -ef shows the command runs from nobody user 2. sudo noip2 -S says it updates every 30 minutes via /dev/eth0 with NAT enabled.

Update: An alternative is to use ddclient. However, ddclient never updates the IP.

Mail

  • Allow only one MX record for each host for free no-ip account.
  • Click Host/Redirects > Manage Hosts > Modify.

webmin

See http://www.webmin.com/deb.html 

wget http://prdownloads.sourceforge.net/webadmin/webmin_1.600_all.deb
dpkg --install webmin_1.600_all.deb

The install will be done automatically to /usr/share/webmin, the administration username set to root and the password to your current root password. You should now be able to login to Webmin at the URL http://localhost:10000/. Or if accessing it remotely, replace localhost with your system's IP address.

Install Virtualbox on ubuntu

See here.

Remote desktop

Remote desktop connection from Windows 7

Install xrdp on Ubuntu. That's it. See 

sudo apt-get install xrdp
sudo /etc/init.d/xrdp start

Remote desktop connection to Windows 7 from xubuntu

sudo apt-get update
sudo apt-get install rdesktop
rdesktop xxx.xxx.x.x -f -u USERNAME -p PASSWORD
rdesktop 192.168.1.4 -g 1280x720 -u USERNAME -p PASSWORD

where -f option means full screen and -g means geometry.

Remote desktop connection to NCI

ts.nci.nih.gov:1494

DOMAIN: NIH

resolution: 1024 x 768

Use remmina or rdesktop or freerdp (sudo apt-get install freerdp-x11) 

rdesktop ts.nci.nih.gov:1494 -d NIH -u XXXXX -g 1024x768

To share a folder from the local machine, use "-r" option 

rdesktop ts.nci.nih.gov:1494 -d NIH -u XXXXX -g 1280x1024 -r disk:remotedisk=/home/$USER/Downloads

Remote desktop connection to Ubuntu

Network -> Remmina. Protocol: VNC-virtual network. Input IP, username and password.

Allow for remote desktop connection

  1. Go to System -> Preference -> Remote desktop. Allow other uses to view your desktop & uncheck you must confirm each access & require user to enter this password.
  2. Go to System -> Preference -> Monitor. Change monitor resolution to 1280 x 720.

Allow for remote desktop connection when vino failed

Try teamviewer. It works fine. After launching it, the software automatically creates an ID and password. We can change the password so it is fixed. Then launch the software on the client. Use the partner's ID and password to connect to it.

Remote desktop connection through vmware workstation

We shall be able to remote desktop connect to a Windows guest machine if the guest machine has configured to use bridge connection and a static IP. If there is a problem, it is likely caused by Window's firewall. See the two screenshots. We can just turn off the firewall of home network but keep the firewall on for the public network.

We don't need to use port forward for the remote desktop connection.

ViewActiveNetwork.png Firewall.png

Instal sshd

apt-get update
apt-get install openssh-server

Install LAMP

See the page [1] 

apt-get install apache2
a2enmod rewrite
apt-get install mysql-server   [I choose branch name as MYSQL root password]
apt-get install php5 php-pear php5-suhosin
apt-get install php5-mysql
/etc/init.d/apache2 restart

Another instruction including how to set up user directories for Apache web server http://wiki.debian.org/LaMp 

apt-get install mysql-server mysql-client
mysql_secure_installation
apt-get install apache2 apache2-doc
apt-get install php5 php5-mysql libapache2-mod-php5
apt-get install python libapache2-mod-python

The apache configuration file is in /etc/apache2/apache2.conf.

For PHP, it is also useful to install php for command line. 

sudo aptitude install php5-cli

Restarting apache before testing on web browser 

/etc/init.d/apache2 restart

Install mediawiki using tar ball

http://www.mediawiki.org/wiki/Manual:Running_MediaWiki_on_Ubuntu

Install mediawiki using aptitude

And also the link http://www.mediawiki.org/wiki/Manual:Running_MediaWiki_on_Debian_GNU/Linux 

aptitude install mediawiki php5 apache2 mediawiki-extensions libapache2-mod-php5

This will install latex. After the end, we can use Synaptic package manager to see what were installed. Now following the instruction in https://help.ubuntu.com/community/MediaWiki, we remove the '#' from the third line so that it reads 'Alias /mediawiki /var/lib/mediawiki': 

sudo nano /etc/mediawiki/apache.conf
sudo /etc/init.d/apache2 restart

Now we can start mediawiki by opening a browser and pointing it to http://localhost/mediawiki.

We need to enter 

Site config
  admin username: WikiSysop
  password:

Database config
  Database name: wikidb
  DB username: 
  DB password: 
  Superuser name: root
  Superuser password: [depend on how it was chosen when installing MYSQL]

Press the button of "Installing mediawiki". We will be welcomed to the wiki page. Following the instruction,

For security reason, I remove new account creation and anonymous editing. I also remove edit counters.

Backup and Restore mediawiki

See docs.google.com note. The process involves 3 parts: mediawiki system, mysql and images.

Install moinmoin

UpnP server

http://bbrks.me/rpi-minidlna-media-server/

It works even I use my phone to tether data (I don't need to turn on wifi on my phone). 

sudo apt-get update
sudo apt-get install minidlna
sudo nano /etc/minidlna.conf # the default location of media files is on /var/lib/minidlna
sudo service minidlna start
sudo update-rc.d minidlna defaults # ask minidlna to start up automatically upon boot.

MYSQL security

Just execute mysql_secure_installation from the command line.

  • You can set a password for root accounts.
  • You can remove root accounts that are accessible from outside the local host.
  • You can remove anonymous-user accounts.
  • You can remove the test database, which by default can be accessed by anonymous users.

See http://www.mysql-optimization.com/mysql-secure-installation-program.html

Compiling R

Use the following command to download required components before building any R packages. See also ubuntu package for R 

sudo apt-get build-dep r-base

PPA management

Package maintenance

https://help.ubuntu.com/community/AptGet/Howto#Maintenance_commands

Show package information

apt-cache show <package_name>

List files in a package

dpkg -L <package_name>

List racing games package (kind of search packages by key words)

apt-cache search racing game

List all packages installed in the system

apt-cache pkgnames

Search installed packages

dpkg -l libgtk* | grep -e '^i'
dpkg -l libpng* | grep -e '^i'
dpkg -l libjpeg* | grep -e '^i'

List of installed packages

dpkg --get-selections

Or using dpkg-query utility. 

dpkg-query -l

Show (sort) package size

dpkg-query -Wf '${Installed-Size}\t${Package}\n' | sort -n

Check if a library is installed or not

Use ldconfig -p | grep LIBNAME. For example, to check if libxml is installed or not, 

EXIST=`ldconfig -p | grep libxml | wc -l`
if [ $EXIST -ng 0 ]; then echo EXISTING; fi

List of available (uninstalled) packages

aptitude -F "%p" search "?not(?installed)"

See http://superuser.com/questions/408082/how-do-i-get-a-list-of-available-i-e-uninstalled-packages-in-debian

Clean up/remove packages

ubuntugeek.com

View logs

  • Apache /var/log/apache2/access.log

Torrent

Popular search sites

  • torrentz
  • thepiratebay
  • isohunt
  • kickass torrents

Torrent client command line: aria2c

aria2 is a lightweight multi-protocol & multi-source command-line download utility. It supports HTTP/HTTPS, FTP, BitTorrent and Metalink. aria2 can be manipulated via built-in JSON-RPC and XML-RPC interfaces.

http://linuxconfig.org/aria2-all-in-one-command-line-download-tool 

sudo apt-get install aria2
aria2c magnet:?xt=urn:btih:1e99d95f1764644a86a8e99bfd80c ...

Torrent client: transmission-cli

https://wiki.archlinux.org/index.php/Transmission

Torrent client command line: rtorrent

We first need to create an .rtorrent.rc file under $HOME directory. Then run 

rtorrent XXX.torrent

In the simplest case, .rtorrent.rc looks like 

# On the terminal, mkdir ~/Downloads/rsession
download_rate = 0
upload_rate =50
directory = ~/Downloads
session = ~/Downloads/rsession
port_range = 55556-55560
scgi_port = 127.0.0.1:5000
use_udp_trackers = yes
encryption = allow_incoming,try_outgoing,enable_retry

As you can see here, I have created a sub-directory rsession under ~/Downloads/.

Keybinding:

  • ctrl + q = quit application
  • ctrl + d = stop an active download
  • ctrl + s = start downloading

Install webmin

apt-get install perl libnet-ssleay-perl openssl libauthen-pam-perl libpam-runtime libio-pty-perl apt-show-versions python
wget http://prdownloads.sourceforge.net/webadmin/webmin_1.600_all.deb
dpkg --install webmin_1.600_all.deb

Webmin.png

File does not exist: /var/www/favicon.ico from /var/log/apache2/error.log

The reason? You have not created a favicon, also known as a website icon, for your website. It’s the icon that displays in the address bar of your web browser when you connect to a website. A web browser will request this icon file from every website.

If you choose to create one. Use a program such as Gimp and create a 16×16 pixel image and save it as a .ico filetype. Then upload that file to the DocumentRoot of your website. You will need one for each VirtualHost. If you don’t have Gimp, there are online resources such as favicon.cc where you can create a .ico file and download it for your own use.

As you know by now, not having a favicon.ico file, won’t stop web browsers from requesting it each time. But you can tell Apache not to log the event as an error message. You will still see the request in the access.log, but at least you will have a cleaner error.log file.

Add the following block of code to each VirtualHost, or at least the ones which don’t have a favicon file. 

Redirect 404 /favicon.ico
<Location /favicon.ico>
 ErrorDocument 404 "No favicon"
</Location>

Don’t forget to restart apache after making the change. If you want make a “global” change, which would apply to any and all VirtualHosts, you can create a file in Apache’s conf.d folder with a name such as nofavicon.conf and then add that block of code to the file. That would disable favicon across the board and save you from having to edit each VirtualHost.

Or, you create an empty file with the name “favicon.ico” in the directory root of Apache (for exemple /var/www/).

Show weather on the taskbar

http://www.noobslab.com/2012/10/important-thingstweaks-to-do-after.html

Screencaster

Watch TV

Freetuxtv

Running concrete5 On Nginx (LEMP)

http://www.howtoforge.com/running-concrete5-on-nginx-lemp-on-debian-squeeze-ubuntu-12.10

Open mms stream in google chrom in linux

http://www.thermetics.net/2011/12/07/how-to-open-mms-links-from-chrome-under-ubuntu/

Download mms stream (suitable if mms is an extension; for example, studioclassroom)

  • Use mimms (will NOT output to speaker at the same time; so is best in terms of performance)
sudo apt-get install mimms
mimms -t 60 mms://example.com/video

where -t option specifies number of minutes. See http://linuxers.org/howto/how-download-mms-streaming-videos-ubuntu The output will have the same extension as the input. But it is not always clear. For example 

$ mimms -t 3 mms://bcr.media.hinet.net/RA000073                                                                                                    
mms://bcr.media.hinet.net/RA000073 => RA000073.wmv
1.48 MB / ∞ B (8.79 kB/s, ∞ s remaining)           
Download stopped after user-specified timeout.


  • Use mplayer (will output to speaker at the same time, so suffer from performance, anyway do not get output)
mplayer mms:/link/something.xxx -dumpstream -dumpfile file.xxx

mms:/link/something.xxx - link to the stream you wish to download file.xxx - file to which you wish to download the stream, be careful to write the same extension xxx

Wait for the file to download and that's it. See http://ubuntuhowtos.com/howtos/download_mms_stream

Keep a linux process running after log out

# nohup command-with-options &

Open Firefox in cron job

The trick is to use export DISPLAY=:0 

export DISPLAY=:0
firefox http://www.google.com &

Close firefox gracefully

Use wmctrl command. 

sudo apt-get install wmctrl
wmctrl -c firefox

Record audio out from command line

http://www.pantz.org/software/alsa/recording_sound_from_your_web_browser_using_linux.html

Step 1: Install required programs 

sudo apt-get install gnome-media pavucontrol lame

Step 2: Create a script file <recordfm.sh> 

#!/bin/bash
# Get pulseaudio monitor sink monitor device then pipe it to 
# sox to record wav, lame to encode to mp3, or flac to encode flac
FILENAME="$1"
STOPTIME="$2"
# Encoding options for lame and flac.
LAMEOPTIONS="--preset cbr 192 -s 44.1" 

if [ -z "$FILENAME" ]; then
    echo -e "
    Usage: $0 /path/to/output.mp3
    Usage: $0 /path/to/output.mp3 stopinseconds" >&2
    exit 1
fi

# Get sink monitor:
MONITOR=$(pactl list | egrep -A2 '^(\*\*\* )?Source #' | \
    grep 'Name: .*\.monitor$' | awk '{print $NF}' | tail -n1)
echo "set-source-mute ${MONITOR} false" | pacmd >/dev/null

# Record it raw, and pipe to lame for an mp3
echo "Recording to $FILENAME ..."

if [[ $FILENAME =~ .mp3$ ]]; then
  if [ -z $STOPTIME ]; then
    parec -d $MONITOR | lame $LAMEOPTIONS -r - $FILENAME 
  else
    echo -e "\nStopping in $STOPTIME seconds"
    parec -d $MONITOR | lame $LAMEOPTIONS -r - $FILENAME 2>&1 &
    SPID=$!
    sleep $STOPTIME
    kill -9 $SPID
  fi
fi

Step 3: play the music or launch a browser with a desired url.

Step 4: run the bash script 

chmod +x recordfm.sh
./recordfm.sh test.mp3 10

where <test.mp3> is the output filename and 10 is recording length (seconds). It works.

Note the script teaches us how to find out the ID for a process we just launched (cleaner than using ps -ef | grep commands). The command is 

SPID=$!
echo $SPID

Use VLC to record internet radio (suitable if the stream is continuous)

sudo apt-get install vlc browser-plugin-vlc

sudo apt-get install ubuntu-restricted-extras

sudo apt-get install lame  libmp3lame0
sudo apt-get install ffmpeg
sudo apt-get install libavcodec-extra-53 libavdevice-extra-53 libavfilter-extra-2 libavformat-extra-53 libavutil-extra-51 libpostproc-extra-52 libswscale-extra-2

A successful run will have an output like 

VLC media player 2.0.3 Twoflower (revision 2.0.2-93-g77aa89e)
[0x2329ca8] dummy interface: using the dummy interface module...
[0x7fac2c007428] mux_dummy mux: Open
[0x7fac2c003598] access_mms access: selecting stream[0x1] audio (39 Kib/s)
[0x7fac2c003598] access_mms access: connection successful
[0x7fac2c003598] access_mms access error: failed to send command
[0x7fac2c005fe8] idummy demux: command `quit'

Play audio using a command line mode

See also my Beaglebone page for a comparison of different possibilities. For VLC, there are 3 interface modes. The following example is to run vlc in a text mode with the ncurses library. 

sudo apt-get install vlc-nox
vlc -I ncurses XXX.mp3

Play youtube using VLC from a command line

See this post 

vlc -I http https://www.youtube.com/watch?v=UlW77conmAc

CVS

CVS server

# Original data dir: /home/mli/Downloads/hmv_arc
# CVS dir on server (CVSROOT): /home/mli/cvsrep
# CVS dir on local:  /home/mli/Downloads/localcvs
# Project name: mycvs
# cvs user name: mli

sudo apt-get install cvs

mkdir ~/cvsrep
export CVSROOT=/home/mli/cvsrep
cvs init
sudo groupadd mycvsgrp
sudo useradd -G mycvsgrp mli # Assume mli is a new user
sudo usermod -a -G mycvsgrp mli  # assume mli is an existing user  
groups mli  # view groups a user is in use
sudo chown -R :mycvsgrp /home/mli/cvsrep # change the group ownership of cvsrep directory to mycvsgrp.
cd ~/Downloads/hmv_arc
cvs import -m "initial" mycvs mli START # import files to CVS repository
                                        # the new subfolder mycvs has owner mli.mli

Note 1. It is OK to use the same CVSROOT for multiple modules/projects since each module/project will be saved under a separate subfolder.

Note 2. The cvs version that I have installed in my ubuntu server is 1.12.13. 

$ cvs -v

Concurrent Versions System (CVS) 1.12.13-MirDebian-6 (client/server)


Quick test to checkout project to the same machine 

cd ~/Downloads
mkdir localcvs
cd localcvs
cvs checkout mycvs

This will create a new subfolder 'mycvs' under ~/Downloads/localcvs.

CVS repository on server has a structure 

cvsrep/CVSROOT
cvsrep/mycvs

CVS sandbox on local machine has a structure 

mycvs/CVS
mycvs/[files1]
mycvs/[files2]

CVS client (ubuntu)

cvs -d :ext:[email protected]:/home/mli/cvsrep checkout mycvs

OR 

export CVSROOT=:ext:[email protected]:/home/mli/cvsrep
cvs checkout mycvs

To add/commit a new file 

cvs add mynewfile
cvs commit -m "my log message" mynewfile

cvs update filename

The file's encoding is correct for 'hmv_.rc' file. 

brb@ubuntu:~/Downloads$ cvs -d :ext:[email protected]:/home/mli/cvsrep checkout mycvs
brb@ubuntu:~/Downloads$ file -bi mycvs/src/hmvUnicode.rc
text/x-c; charset=utf-16le

CVS client (windows)

I use WinCVS for a demonstration

  • Remote -> checkout module
  • Module name: mycvs
  • CVSROOT:
    • protocol: ssh
    • repository path: /home/mli/cvsrep
    • user name:
    • host name: taichi.selfip.net

If I use TortoiseCVS (1.12.5 from 1/24/2011), I need to choose ext as protocol instead ssh. Still the checked out file 'hmv_.rc' still contains unreadable Chinese characters. The cvsnt is the latest free version (2.5.05). If I want to use WinCVS + cvsnt from TortoiseCVS, the options in the CVSROOT dialog looks weird and cannot create a connection.

For the unicode encoding. If I commit the file at first from ubuntu os, but check out in Windows. The checked out file has right encoding (using Notepad ++, or from VS2010). However, the file does not have right line ending and it shows Chineses character when I open it in either Notepad++ or VS2010.

To see hidden characters in Linux, try either one of the following 2 methods:

  • Open the file in EMACS and do a M-X hexl-mode
  • geany editor.

The solution I have found to overcome accessing unicode (utf-16) file on Windows OS is using Cygwin.

  • Download setup.exe from http://cygwin.com/install.html
  • Root directory = c:\cygwin
  • Local package directory = C:\Users\brb\Downloads
  • Direct connection
  • Download site: ftp://cygwin.mirrors.pair.com (Some mirrors are not updated & contain old version of packages! For example, make sure the cvs version is 1.12.13.)
  • Search: cvs. Click plus sign next to "Devel". Click 'Skip' corresponding to cvs package.
  • Search: ssh. Click plus sign next to "Net". Click 'skip' correspond to openssh package.
  • Click 'Next' button.
  • Click 'Finish' button.
  • Now open 'Cygwin Terminal' icon on Windows Desktop.
export CVSROOT=:ext:[email protected]:/home/mli/cvsrep
cvs checkout mycvs

The 'mycvs' directory should be under C:\cygwin\home\brb (a.k.a. /home/brb in cygwin) directory. We can open 'hmv_.rc' file in Notepad++ to double check if the file looks normal OR use md5sum to check.

Difference between CRLF (Windows), LF (Linux, Mac) and CR

This is a good summary I found: http://stackoverflow.com/questions/1552749/difference-between-cr-lf-lf-and-cr-line-break-types

  • The Carriage Return (CR) character (0x0D, \r) moves the cursor to the beginning of the line without advancing to the next line. This character is used as a new line character in Commodore and Early Macintosh operating systems (OS-9 and earlier).
  • The Line Feed (LF) character (0x0A, \n) moves the cursor down to the next line without returning to the beginning of the line. This character is used as a new line character in UNIX based systems (Linux, Mac OSX, etc)
  • The End of Line (EOL) character (0x0D0A, \r\n) is actually two ASCII characters and is a combination of the CR and LF characters. It moves the cursor both down to the next line and to the beginning of that line. This character is used as a new line character in most other non-Unix operating systems including Microsoft Windows, Symbian OS and others.

Common CVS commands

cvs checkout MODULE
cvs checkout DIR
cvs checkout DIR/SUBDIR
cvs co DIR/SUBDIR/FILENAME
# check out a specific tag and put it in a specified directory. 
# the specified directory name will replace the module name in output.
mkdir localcvs
cvs checkout -r v4_3 -d localcvs MODULE

cvs add myfile.c
cvs add -kb myfile.bin
# If you accidentally add a file, simply skip the commit for that file.

cvs update –dA DIR/SUBDIR 
# -d: Create  any  directories  that  exist in the repository if they're missing from the working directory.
# -A: Reset any sticky tags, dates, or -k options. Needed after you use "cvs update -D" or "cvs update -r".
cvs update –A DIR/SUBDIR/FILENAME    

cvs commit
cvs commit –m "add test suite" DIR/SUBDIR/FILENAME

mkdir ~/original
touch ~/original/newfile
cvs import ~/original VENDORTAG RELEASETAG

mkdir ~/localcvs
cd ~/localcvs
cvs checkout common/too

cvs diff -r1.23 -r1.24 SUBDIR/FILENAME     # Difference between specified versions 1.23 & 1.24.
cvs diff -D "1 hour ago" MODULE
cd LOCALCVS; cvs diff 
cvs checkout -D "1 hour ago" MODULE
cvs checkout -D "2013-02-27 01:30" MODULE

rm file(s); cvs remove file(s); cvs commit -m "Comment text" file(s)
# You must rm the file before issuing the cvs remove command. The remove is not final until the commit has been issued.

# cvs does not let you remove directories. However it does let you ignore any directories that are empty.
cvs co -P Offline
cvs update -PdA
# P: Prune empty directories. d: create new directories. A: Reset sticky tags

cvs history -c -a -D "1 day ago"  -z "-0500"
# find all changes submitted to the repository by anyone in the past day. 
# -z is used to adjust the time zone.

cvs history -c -a -D "1 day ago" -f Mu2eG4/src
# find all changes submitted to Mu2eG4/src (or any other subdirectory) by anyone in the past day

cvs history -c -u USER "1 day ago" -f Mu2eG4/src 
# find all changes submitted by USER to Mu2eG4/src in the past day

cvs log FILENAME

Bypass SSH password login (convenient for CVS, git etc)

http://www.howtogeek.com/tips/bypass-ssh-logins-by-adding-your-key-to-a-remote-server-in-a-single-command/

  1. ssh-keygen -t rsa
  2. (make sure the remote server has .ssh directory)
  3. cat ~/.ssh/id_rsa.pub | ssh user@hostname 'cat >> .ssh/authorized_keys'
  4. ssh user@hostname

It helps with CVS log in too when the CVS works by using ssh protocol. Note that step 3 allows to run a shell command at a remote machine.

See https://help.github.com/articles/generating-ssh-keys also for similar instruction when work on github.

The ssh key can be copied to another a machine (pay attention to mode). Or let the new machine to create its own key pair and use ssh-copy-id to append the identity file to remote machine's ~/.ssh/authorized_keys file. See http://superuser.com/questions/332510/how-to-transfer-my-linux-ssh-keys-to-another-machine.

We can even have multiple ssh key on local machine by using. ssh/config file. See http://www.karan.org/blog/index.php/2009/08/25/multiple-ssh-private-keys.

Install Ubuntu to a USB flash drive

The following approach assumes the boot loader will not be put on internal hard disk if you are careful enough. See also http://fernhilllinuxproject.com/guidesandhowtos/installubuntutousbdrive.html

  1. Boot from live CD in ubuntu.
  2. Insert the USB flash drive in USB Port.
  3. On desktop double click on icon 'install ubuntu 12.04'
  4. Click on continue and it will ask if you want to unmount the flash drive click on yes
  5. Choose some thing else when ask to where to install ubuntu.
  6. Next you will see your hard disk partitions and flash drive. Click on flash drive partition and then click on change. Change the partition type to ext4 and mount as to '/'. Click on ok to close the dialogue box.
  7. In last you will see a drop down menu on where to install the the boot loader. Initially it will be showing sdc,sdd but on drop down you must select sdc1 or sdd1. Failing this step you may render your computer unbootable.
  8. Then click on install and linux will be installed on your USB Flash drive.

Install a new hard drive

https://help.ubuntu.com/community/InstallingANewHardDrive

  1. Use sudo fdisk command to create partition table. Then 'n', 'p', '1', 'w' and several returns.
  2. Use sudo mkfs -t ext4 /dev/sdb1 to create a new partition
  3. Use sudo mkdir /mnt/ssd to create a new mount point
  4. Use sudo nano -Bw /etc/fstab to do auto mount on boot
  5. sudo chown -R USERNAME:USERNAME /mnt/ssd to give the ownership to the USERNAME.

Building a simple Beowulf Like Cluster with Ubuntu

MPICH2

Resource:

  1. http://byobu.info/article/Building_a_simple_Beowulf_cluster_with_Ubuntu/
  2. https://help.ubuntu.com/community/MpichCluster
  3. https://help.ubuntu.com/community/SettingUpNFSHowTo (NFS configure)

Here is my record for creating a cluster environment based on ubuntu 13.04. The master node is running on ubuntu 13.04 desktop with virtualBox 4.2. The virtualBox has added a host-only adapter (vboxnet0) with ip 192.168.56.1. This adapter will be added to the master node so I can use this ip to identify the master node in the host-only network. Creating cluster using VirtualBox is just for the education purpose, not for real practice.

  • (virtualBox) Create two virtual machines running on ubuntu 13.04 server. The ssh server was checked during installation. The host name for each of them is ubuntuNode1 and ubuntuNode2 respectively. The network adapter is left by default (NAT) during installation. But after the installation is done, I shutdown the system and add a host-only adapter (vboxnet0) to each of them. Then after the system is up again, I change the IP so it is static. Do sudo nano /etc/network/interfaces and append the following before running sudo /etc/init.d/networking restart to take the change in effect.
auto eth1
iface eth1 inet static
address 192.168.56.101 # use 192.168.56.102 for ubuntuNode2
netmask 255.255.255.0
network 192.168.56.0
broadcast 192.168.56.255

Note that it is better not to add host-only network before installation, or during installation it will ask what is the primary network (confusing). So at the end each new nodes should have both eth0 and eth1 adapters and they should not reside in same subset. ifconfig may not show all adapters so we should use ifconfig -a instead. Although host-only network is used for communication between guest and guest OR guest and host, since each guest node has NAT adapter by default so the guest machine can still access the internet.

  • (master node) Edit /etc/hosts so it like like
127.0.0.1 localhost
192.168.56.1 ubuntu1304
192.168.56.101 ubuntuNode1
192.168.56.102 ubuntuNode2

Note that the master node will be used to start jobs on the cluster although it is OK to let the master node as one of nodes to execute the jobs.

  • (all nodes) Run
sudo adduser mpiuser --uid 999

It is recommeneded to use the same password for the user. This will create a new directory /home/mpiuser. This is the home directory for user mpiuser and we will use it to execute jobs on the cluster.

  • (master node) Run
sudo apt-get install nfs-kernel-server
  • (other nodes) Run
sudo apt-get install nfs-client
  • (master node) Add the following to the file /etc/exports
/home/mpiuser *(rw,sync,no_subtree_check)

or something like 

/home/mpiuser 192.168.56.0/24(rw,sync,no_subtree_check)

Some people create a shared folder under /srv directory.

Now run 

sudo service nfs-kernel-server restart
  • (master node)
sudo ufw allow from 192.168.56.0/24
  • (other nodes)
sudo mount ubuntu1304:/home/mpiuser /home/mpiuser

And if we want to mount the NFS shared directory when the compute nodes are booted, edit /etc/fstab by adding 

ubuntu1304:/home/mpiuser /home/mpiuser nfs
  • (master node)
sudo apt-get install ssh
su mpiuser
ssh-keygen
ssh-copy-id localhost

We can test if the ssh works without passwords 

ssh ubuntuNode1
echo $HOSTNAME
  • (all nodes)
sudo apt-get install mpich2
which mpirun
which mpiexec
  • (master node, mpiuser)

Go to the home directory of mpiuser and create a new file hosts. Include host names for computing nodes (it is OK to include master node, ubuntu1304, as one of computing nodes) 

ubuntuNode1
ubuntuNode2
  • (all nodes)
sudo apt-get install build-essential
  • (master node, mpiuser). Creating a test file <mpi_hello.c>
#include <stdio.h>
#include <mpi.h>

int main(int argc, char** argv) {
    int myrank, nprocs;

    MPI_Init(&argc, &argv);
    MPI_Comm_size(MPI_COMM_WORLD, &nprocs);
    MPI_Comm_rank(MPI_COMM_WORLD, &myrank);

    printf("Hello from processor %d of %d\n", myrank, nprocs);

    MPI_Finalize();
    return 0;
}

And compile and run it by 

mpicc mpi_hello.c -o mpi_hello
mpiexec -n 3 -f hosts ./mpi_hello

We should see something like 

Hello from processor 0 of 2
Hello from processor 1 of 2

Successful! That's it.

OpenMPI

With R

Sharing internet via wifi

I cannot make it to work to share the internet via wifi on my Xubuntu 13.04. However the solution in http://askubuntu.com/questions/287251/creating-an-infrastructure-hotspot-using-ubuntu-12-10 works for me. 

$ sudo su -
# add-apt-repository ppa:nilarimogard/webupd8
# aptitude update
# aptitude install ap-hotspot
# ap-hotspot configure
# ap-hotspot start

In the step of 'ap-hotspot configure' it will ask for Access Point name and WPA passphrase.

When I run the final line 'ap-hotspot start', it will ask me to disconnect my current wifi first.

At the end, I found my original configuration in ubuntu does not work even its setting is there. apt-hotspot create an infrastracture AP instead of adhoc.

~/.xsession-errors file is filling the hard disk

It is related to vino-server. See

Short solution is to use "kill -9 xxx" to kill the process and rm to remove ~/.xsession-errors file. The long time solution is to uninstall vino.

JRE and JDK

Install openjdk or Sun jdk. See http://www.maketecheasier.com/install-java-runtime-in-ubuntu/ (Ubuntu 12.04)

If we have multiple versions of JRE/JDK, we can use the following command to set the default version 

sudo update-alternatives --config java

This approach seems to be working in the case JAVA_HOME cannot be honored.

Oracle JAVA

http://askubuntu.com/questions/521145/how-to-install-oracle-java-on-ubuntu-14-04 

sudo apt-add-repository ppa:webupd8team/java
sudo apt-get update
sudo apt-get install oracle-java8-installer

java -version

CPU/system load

  • CPU-G: See this instruction to install CPU-G program for viewing hardware information (process, mb, graphic, memory, system).
sudo apt-get install indicator-multiload

Unlock keyring

I got the prompt of unlocking keyring every time I open google chrome browser. 

google-chrome --password-store=basic

It is also helpful to remove ubuntu one from Ubuntu 12.04. See http://hex.ro/wp/blog/removing-ubuntuone-from-ubuntu-12-04/

netcat (nc) - arbitrary TCP and UDP connections and listens

Netcat or nc is a networking utility for debugging and investigating the network.

The nc (or netcat) utility is used for just about anything under the sun involving TCP, UDP, or UNIX-domain sockets. It can open TCP connections, send UDP packets, listen on arbitrary TCP and UDP ports, do port scanning, and deal with both IPv4 and IPv6.

nmap - port scanning

nmap - Network exploration tool and security / port scanner 

sudo apt-get install nmap
nmap -T4 -F 192.168.1.99-255

sudo nmap -sS -O 192.168.1.99
nmap 192.168.1.99    # does not require root privileges

List of all services/daemons

Run service --status-all to get a list off all the Upstart services and their status. See

  • man service
  • man initctl
service --status-all   # output format is clean

sudo initctl list      # show the process number too

where in the output "+" means started, "-" stopped, and "?" unknown.

Google Drive

sudo add-apt-repository ppa:thefanclub/grive-tools
sudo apt-get update
sudo apt-get install grive-tools

On Dash, type grive. It will ask you to enter a sequence of characters. After that, a new folder 'Google Drive' will be created and files will be downloaded.

Video rip/convert/transcoder

sudo add-apt-repository ppa:stebbins/handbrake-releases
sudo apt-get update
sudo apt-get install handbrake-gtk

sudo apt-get install handbrake-cli
  • libav-tools

To convert youtube flv file to mp4. 

sudo apt-get install libav-tools
avconv -i INPUT.flv -codec copy OUTPUT.mp4

To extract audio only: 

avconv -i INPUT.flv -codec copy -vn OUTPUT.mp4

Video editing in Linux

Blender

Kdenlive

OpenShot

An article from PC World. 

sudo apt-get install openshot

When I needed to export the video (choose 'youtube' profile, 'youtube-HD' target, 'HD 270, 29 .97 bps', and 'high' quality), I found I need to install libx264 code. On Ubuntu, I open software center and seach 'libavformat'. I choose 'libavformat-extra-53'.

VideoLAN Movie Creator

ffdiaporama

Create videos from images, movie clips and music.

Youtube command line tools

  • Download youtube video
sudo apt-get install youtube-dl
  • For videos with 1080p resolution, youtube-dl will download 720p by default. To download 1080p, see this post
youtube-dl -f 137+141 XXXX
  • To convert the video to mp3 format, use soundconverter
sudo apt-get install soundconverter

mps-youtube
  • 4k videos samples (vlc v2.1.4 cannot play)

Unity LauncherSwitcher

Ubuntugeek.com

Create Unity Launcher

Take RStudio for example,

  • Desktop icons are saved under /usr/share/icons/hicolor/16x16/apps/rstudio.png, where other than 16x16 there are also icons from 24x24, 32x32, 48x48 sizes (directories). Another place is ~/.local/share/icons/hicolor/16x16/ .
  • Unity launcher is created at /usr/share/applications/rstudio.desktop or ~/.local/share/applications/XXXX.desktop.
  • Desktop shortcut can be created by copy XXXX.desktop to ~/Desktop directory.

The icon size in the launcher can be adjusted by opening System Settings -> Apperance -> Launcher Icon Size (at the bottom).

k2pdfopt has an example how to do it for creating a desktop shortcut and be in the right click menu.

How to Convert the Unity Launcher into a Dock-Style Launcher

See an article from Howtogeek.com.

What date was the system installed

ls -l /var/log/installer

What is the last log in time for users

lastlog  # all users
last     # current user

What is the reboot time

last reboot

Crop an image

sudo apt-get install gthumb

Open an image. Click color palette icon on top-right corner (or use keyboard shortcut 'e'). Click 'crop'. There will be a rectangle on image where you can resize the rectangle.

Mind-mapping

KeePass

Search Ubuntu Software Center to install KeePassX (more reviews) or KeePass2. There is no need to use ppa.launchpad.net to install from. Note that the kdbx file used in KeePass2 cannot be opened in KeePassX. To use the command line to install KeePass2, 

sudo apt-get install keepass2

Note:

Update Firefox

See https://help.ubuntu.com/community/FirefoxNewVersion. The following instruction is used to get security-testing packages. 

sudo add-apt-repository ppa:ubuntu-mozilla-security/ppa
sudo apt-get update
sudo apt-get install firefox

Change default audio player

Right-click an MP3 file, and from the menu select “Properties”. In the window that appears, go to the “Open With” tab and select VLC or whatever. Click the “Set As Default” button to set VLC as the default player.

You might also want to set the default player in the sound menu.

Unity display timeout

Go to Dash -> power setting -> Brightness Settings

Screensaver

I like gluqlo (flip clock) screensaver. However, it is not easy to make it to replace the gnome-screensaver. 

sudo apt-add-repository ppa:alexanderk23/ppa
sudo apt-get update
sudo apt-get install gluqlo
  • Follow the instruction to install it on Ubuntu machine.
  • Follow the instruction there to remove gnome-screensaver and install xscreensaver.
  • Configure xscreensaver to use only 1 screensaver. Edit ~/.xscreensaver file and add a line like
gluqlo -root                                \n\
  • Still follow the instruction to allow xscreensaver to start when the machine starts up. Don't try to edit ~/.xinitrc file as other sites suggested; start Dash and type 'startup' and follow the screen dialog to add xscreensaver -nosplash.
  • Still follow the instruction to add lock screen keyboard shortcut.

Note that the above steps work for Ubuntu 12.04 & 13.10 but not 14.04 (Ubuntu 14.04 changed to use LockScreen instead LightDM program to lock the screen). A solution on Ubuntu 14.04 is to disable screen lock.

  1. Still follow the above instruction to remove gnome-screensaver and install xscreensaver. Set gluqlo as the only one screensaver.
  2. Go to Brightness & Lock panel from the Unity Launcher. And set Turn screen off when inactive: to Never.
  3. Install "Unity Tweak Tool" with sudo apt-get install unity-tweak-tool. Run it from the Launcher and select System > Security > Enhance system security by disabling Desktop lock.
  4. To enable locking desktop (required a password to unlock the desktop), check 'Lock Screen After' option. I pick 1 minute.

For some reason, gluqlo suddenly uses all my cpu (6 cores) resource (Ubuntu 14.04). The computer thus makes some noise. I have to use 'kill' command to kill them.

Flash for browser

On Ubuntu 12.04, there is no way to directly install the pepperflashplugin-nonfree plugin. So we have to use ppa from other people. See this post 

sudo apt-get update
sudo apt-get install chromium-browser
sudo add-apt-repository ppa:skunk/pepper-flash
sudo apt-get update
sudo apt-get install pepflashplugin-installer
sudo update-pepperflashplugin-nonfree --install

The last step gives me an error: sudo: update-pepperflashplugin-nonfree: command not found

Gedit

  • Restore tabs plugin. It works on my gedit 3.4 (ubuntu 12.04). Follow the instruction there exactly.
  • Source code browser plugin. This makes gedit a good IDE for developing C++/Java code since the left panel can show symbols. Click F9 to show the side panel.
  • Darkmate theme.
cd /usr/share/gtksourceview-3.0/styles
sudo gedit darkermate.xml

gedit > Edit > preferences > font and colors > color scheme.

HTML editor

  • Kompozer and the installation instruction for Ubuntu.
  • BlueGriffon: a new WYSIWYG content editor. The interesting thing is the software BlueGriffon EPUB Edition: a cross-platform Wysiwyg editor able to natively create and edit EPUB2 and EPUB3 ebooks!

chm reader

sudo apt-get install xchm

SCR3310 smart card

  • Search "SCR3310 driver linux" on google.com.
# Download pcsc-lite-1.8.13.tar.bz2 from 
# https://alioth.debian.org/frs/?group_id=30105
sudo apt-get install libudev-dev
cd pcsc-lite-1.8.13
./configure
make
sudo make install

# Download libusb
http://libusb.info/
cd libusb-1.0.19
./configure
make
sudo make install

# Download scmccid_5.0.27_linux
# http://www.identive-infrastructure.com/index.php/products-solutions/smart-card-readers-a-terminals/smart-card-readers/scr3310

cd scmccid_5.0.27_l32r
sudo ./install.sh

Chroot

Note that we have to change the conf file a little bit. The 'location' word needs to be changed to 'directory'. Also at the last step when we are ready to test a 32-bit GUI app, we need to issue DISPLAY in a separate line; such as 

export DISPLAY=:0.0
su brb # brb is my root user in the host system that can invoke the schroot program
       # firefox does not allow to use root to start it
firefox

For a recap: 

1. Install the packages
sudo apt-get install debootstrap schroot -y
2. Create a schroot configuration file
sudo nano /etc/schroot/chroot.d/precise_i386.conf
3. Install 32-bit ubuntu with debootstrap
sudo mkdir -p /srv/chroot/precise_i386
sudo debootstrap --variant=buildd --arch=i386 precise /srv/chroot/precise_i386 http://archive.ubuntu.com/ubuntu/
4. Test the chroot environment
schroot -l
schroot -c precise_i386 -u root
uname -a
cat /etc/issue
5. Additional configuration
apt-get install ubuntu-minimal
# That's all.

The article also mentioned the home directories (Documents, Downloads, ...) of the users within the chroot are shared with the host. How to access them from the host?

Check/Diagnostic SD card

http://askubuntu.com/questions/69932/is-there-an-sd-card-diagnostic-utility

Look for Disk Utility on you dash (Alt+F2 and type 'disk')

Ubuntu Snappy Core

How Snappy packages are different from Deb

An article from PCWorld

  • Applications are no longer installed system-wide. The base Ubuntu operating system is kept securely isolated from applications you install later. Both the base system and Snappy packages are kept as read-only images.
  • Snappy packages can include all the libraries and files they need, so they don’t depend on other packages.
  • An update can never fail, as a package installation could potentially fail and become incomplete with typical Linux packages.
  • Snappy also supports “delta” updates, which means only the changed bits of the package need to be downloaded and installed.
  • Snappy-based Ubuntu systems might be standard.

Docker

I haven't found any tutorial yet!

Ova image

Sorry, I don't get the command line back. Booting stuck in the middle.

Beaglebone

http://beagleboard.org/snappy or http://www.ubuntu.com/things#try-beaglebone

Raspberry Pi 2

https://darrenjw2.wordpress.com/2015/02/07/getting-started-with-snappy-ubuntu-core-on-the-raspberry-pi-2/

unable to open mtp device ubuntu 14.04

Use this tip. 

sudo apt-get install mtpfs

RHEL/CentOS

Change hostname

  • Change the ^HOSTNAME line in /etc/sysconfig/network
  • Change the hostname in /etc/hosts
  • Run /bin/hostname new_hostname for the hostname change to take effect immediately.
  • Run /sbin/service syslog restart for syslog to log using the new hostname.

Note that using the command line 'hostname' to change the machine's hostname works only for the current session.

switch to root

su   # Press 'Enter'. It will ask for root's password.

sudoer

Some distributions do not come with sudo command.

As root type:

visudo

and add a line 

MyUserName ALL = ALL

Add an existing user to have sudo privilege

sudo adduser USERNAME sudo

See help.ubuntu.com.

What is my IP address

ifconfig eth0

What is my DNS server

/etc/resolv.conf

What services get started at boot time

chkconfig --list

Is xxx service running

xxx status

What services are currently running

ps -e

and 

lsof -i

will show you services that are listening to TCP or UDP endpoints.

What network ports are open

lsof -i

or 

netstat -aut

What firewall rules do I in place

iptables -L

What packages do I have installed

rpm -qa | less

or 

rpm -qa | grep xxx

What version of package xxx do I have installed

rpm -qi xxx

List of available (uninstalled) packages

yum list available

List All Configured Repositories

yum -v repolist
yum -v repolist | less
yum repolist

To list only enabled repositores 

yum repolist enabled

To list only disabled repositories 

yum repolist disabled

To list available packages under a repo called ksplice-uptrack, enter: 

yum --disablerepo="*" --enablerepo="ksplice-uptrack" list available

yum equivalent of apt-get update

yum check-update

Open a port

Edit the file /etc/sysconfig/optables. Create a new rule by following the examples of other rules in the file.

sendmail

su                             # type your root password to switch the account
yum install m4 telnet mailx
yum install sendmail sendmail-cf
nano /etc/mail/sendmail.mc
m4 /etc/mail/sendmail.mc > /etc/mail/sendmail.cf
service sendmail restart
netstat -an | grep :25 | grep tcp
ps -ef | grep -v grep | grep -i sendmail
nano /etc/mail/local-host-names
service sendmail restart
chkconfig sendmail on
useradd testuser1
useradd testuser2   
passwd testuser2 
mail -s "Test mail from testuser1" testuser2
tail /var/log/maillog
su testuser2 # run 'mail' command to see if the mail has been received.

nano /etc/mail/local-host-names # create a line, says, xyz.com
nano /etc/mail/sendmail.cf      # After the line of "Smart" relay host (may be null), edit as the following
                                # DSmailfwd.nih.gov
nano /etc/postfix/main.cf #  change inet_protocols from all to ipv4.
nano /etc/sysconfig/sendmail    # make sure DAEMON=yes
nano /etc/mail/relay-domains    # this is a new file with 1 line 128.231.90.107
service sendmail restart
mail -s "Test mail from testuser1" [email protected]
tail /var/log/maillog           # Should not see any ERR.
netstat -nutlap | grep 25

Power Manager for GNOME

The configuration defaults for GNOME power manager have not installed correctly. Cannot login

This error will results in a log-in problem except root account. The symptom is 50GB in root (/) is used up.

The problem was caused by a bug in yum where /var/cache/yum/x86_64/6Workstation takes about 42GB space. The 'yum' does not remove old generated .sqlite files.

See https://bugzilla.redhat.com/show_bug.cgi?id=632391

I use 'du -k' command to find out which directory took space. I use 'rm' command to delete the contents.

Even I delete the content, the directory still grows up daily.

Upgrade Python from 2.6.x to 2.7.x

This instruction tells how to install Python 2.7 from source. 

yum -y update
yum groupinstall -y 'development tools'
yum install -y zlib-devel bzip2-devel openssl-devel xz-libs wget
wget http://www.python.org/ftp/python/2.7.8/Python-2.7.8.tar.xz
xz -d Python-2.7.8.tar.xz
tar -xvf Python-2.7.8.tar

# Enter the directory:
cd Python-2.7.8

# Run the configure:
./configure --prefix=/usr/local

# compile and install it:
make
make altinstall

# Checking Python version:
[root@nicetry ~]# python2.7 -V
Python 2.7.8

wget --no-check-certificate https://pypi.python.org/packages/source/s/setuptools/setuptools-1.4.2.tar.gz

# Extract the files:
tar -xvf setuptools-1.4.2.tar.gz
cd setuptools-1.4.2

# Install setuptools using the Python 2.7.8:
python2.7 setup.py install
curl https://raw.githubusercontent.com/pypa/pip/master/contrib/get-pip.py | python2.7 -
pip2.7 install virtualenv

Install Meld

Have not found a solution yet. We need to install it from source. However, the source depends on

  • Python 2.7 (see above for the instruction)
  • GTK+ 3.6
  • GLib 2.34
  • PyGObject 3.8
  • GtkSourceView 3.6

(Update) A binary version of meld is already available in the git. See this post. 

$ cd ~/Downloads/
$ git clone https://git.gnome.org/browse/meld
$ cd meld
$ sudo ln -s /home/$USER/Downloads/meld/bin/meld /usr/bin/meld

VirtualBox guest addition

Check out this post. 

su
# click VirtualBox -> Devices -> Install guest addition
mkdir /media/VirtualBoxGuestAdditions
mount -r /dev/cdrom /media/VirtualBoxGuestAdditions
rpm -Uvh http://dl.fedoraproject.org/pub/epel/6/i386/epel-release-6-8.noarch.rpm
yum install gcc kernel-devel kernel-headers dkms make bzip2 perl
KERN_DIR=/usr/src/kernels/`uname -r`
export KERN_DIR
cd /media/VirtualBoxGuestAdditions
./VBoxLinuxAdditions.run

(Update for 64-bit CentOS 6.5 + VirtualBox 4.3.18) The installation still failed and it showed a missing package which can be installed with 

yum install kernel-devel-2.6.32-431.el6.x86_64

Then I re-run ./VBoxLinuxAdditions.run to finish the installation of guest addition. Reboot and GA works.

Online Tools

Retrieved from "https://wiki.taichimd.us/index.php?title=Linux&oldid=6809"