Enabling the root account is generally considered to not be a good idea in the GNU/Linux desktop world. Using the graphical desktop as root is even more frowned upon. It’s very easy to mess up file permissions, among others, and is very insecure. There are many fine articles on the Internet explaining the problems with this, and if you are considering enabling this account on your day-to-day machine, I strongly recommend you read up on it and/or ask about it on the Ubuntu Forums before going ahead.
That said, sometimes you just want to mess with a system. Try and break it and fix it. Whatever- it’s your system and you can do whatever you like with it.
Note that I take no responsibility for any results you may experience if you follow the instructions given here. Your machine is your responsibility, and you owe it to yourself and your computer to use it responsibly.
Ubuntu comes with the root account disabled. You can use sudo or gksudo for commands that require root permissions. However, enabling root is very simple. Open a terminal, and type
sudo passwd root
You will be prompted to give root a new password.
To log in to your machine as root at startup via the GUI login manager (Lightdm) you need to edit its config file. This is located at /usr/share/lightdm/lightdm.conf.d/50-ubuntu.conf. Edit it with nano (or your preferred editor) by typing the following at the command prompt
sudo nano /usr/share/lightdm/lightdm.conf.d/50-ubuntu.conf
Add this line to the end of the file.
Save the file and restart your computer. The login screen will now let you specify a user name. Enter root, and give the password. Your root account is now up and running.
This post will detail a bit on using the Stockfish chess engine to analyze games with Scid vs. PC.
First step is obviously to install stockfish; in any Debian or Ubuntu based distro this is easily accomplished with
apt-get install stockfish
Next step is to make Scid aware of it. Open up Scid and select Tools>Analysis Engines. Press the “New” button. In the “Configure Engine” dialog, give it a name (Stockfish) a command (which would simply be “stockfish”, no quotes) and set the directory to ~/.scdvspc (by clicking the button labeled such to the right). As Stockfish is a UCI engine make sure the UCI button is selected under Protocol. Click OK and you’re good to go. You can start the engine analyzing the current game selected in Scid from that same dialog by clicking “Start”.
If you’ve started the analysis before you’ve clicked through any moves in the game, it will begin analyzing the entire game. Unless you have it add annotations to each move as it goes, this isn’t very useful. Switch to the Analysis pane in the main Scid window, and press the Annotate button (looks like a page with a pencil and bookmark). The menu that it brings up has a lot of options; seconds to spend on each move or depth per move, blunder threshold, etc. I set the blunder threshold to 0.5 (any move that causes the loss of .5 of a pawn gets flagged), select “All moves” under Add Scores, and set the seconds per move to anything from 30 to 120. These are just some settings I’m experimenting with. You can vary yours as you see fit.
This post is just a few notes on one of the more strangely named programs available for GNU/Linux- not that Linux programs tend to have strange names or anything. 🙂
Scid vs. PC (SVPCC for this article’s purposes) is a fork of the original Scid project. Scid is an acronym for Shane’s Chess Information Database. It is a program for viewing, annotating and analyzing chess games, for playing chess against engines and on online servers, and a database to manage all your games. It is large and complicated and very powerful- sort of the Emacs of chess software. In accordance with my opinions on Emacs, however, I find it well worth learning to use.
I started with the original Scid, but found its interface painfully antiquated. It did more or less function, but I later learned that it has not received any development or bug attention for several years. SVPC doesn’t have the most modern looking interface, but it is very much improved and very usable.
SVPC is easy to install even though all that is provided is the source code. Instructions are available on the project’s website; a simple ./configure and make install. I received errors about no tcl.h or tk.h but a simple installation of tcl8.5-dev and tk8.5-dev fixed that.
I’m currently using Linux Mint 15 with Cinnamon, and I had to create my own .desktop file for Scid vs. PC to access it via the menu or panel, but that was a breeze and there are lots of resources on that subject.
For more Scid vs. PC information there is a good thread on the chess.com forums- link.
I recently undertook the task of building a LFS (Linux From Scratch) installation.
LFS is widely regarded as the ultimate, well, “from scratch” Linux distro. So much so in fact, that it’s really not a distro- it’s a book.
Building an LFS system begins with a host Linux system. A new partition must be created for the LFS system, and source code for GCC, Coreutils, Binutils, and several other programs, to create a toolchain on the new partition. After that, you must compile the tools again (Pass 2) so that you are able to compile the final system against the new kernel headers, and linked to binaries on the new filesystem (that you also create). Then it’s time to write bootscripts and config scripts; configure and compile the kernel, install GRUB (if applicable) and boot into the new system! Seems simple enough. 🙂
I used LFS 7.3 (stable), and built it in a VirtualBox VM with Debian 7 (stable) as the host. I found LFS to be a fairly long and challenging build. It took me about a week all told, and I probably have about 20 hours invested in the system. The main part of it is the endless compiling; I was lucky to be doing this with a recent Intel i7 processor and therefore the actual compiling never went for more than a couple hours. However the book is very clear and gives correct config and make options for each package and I never encountered an error that wasn’t directly related to a typo of my own.
A very useful reference that isn’t emphasized much in the book are the LFS Build Logs. These are the STDOUT logs for each compilation exercise, and show what normal output should look like. On a couple packages I noticed errors after compilation that I found I was able to ignore after checking the logs (relief!). It is absolutely essential that each package compile successfully. The key to a successful installation is being über-OCD.
As it turned out for me, the most difficult part was configuring a bootable kernel. Locating correct modules for my hardware was a bit of challenge; but after the third attempt I was greeted with a login and a bash prompt. 🙂
Not real pretty, not real shiny, but, well, real. 🙂
I plan to make several clones of the functioning LFS system and from there try adding a few things such as X (per instructions in the Beyond Linux From Scratch book).
To anyone considering building LFS; I would definitely recommend it! It was my first time doing much in the way of compiling at all; and my first source-based distro. The trick is just to take it slow; focus; and NEVER skip a step.
Out of pure curiosity, I recently downloaded and installed Scientific Linux 6.4 in a virtual machine. My last extended brush with RPM based distros (besides the couple months I spent with Fedora on my netbook) was back around 2005-2006, when I first experimented a bit with GNU/Linux, specifically Fedora and OpenSUSE.
SL was very remniscent of that experience. It appears to be a virtually identical free clone of Red Hat Enterprise Linux, the de facto corporate distro. It is known mainly for stability, which usually also means “being old”. Both Red Hat and SL use the Gnome2 desktop environment (one that’s been deprecated for nearly a year now) and kernel 2.6. Thus SL is not for people interested in the latest packages or bleeding edge software (obviously). Its server counterpart, CentOS, is also known for stability, and is much used in the server world.
I did install the VirtualBox Guest Additions, as I do with any virtual machine that I plan to use for more than thirty minutes. It was a bit of a challenge, but I did manage to get shared folders/bidirectional copy-paste/arbitrary screen resizing working.
I did have to install gcc, make, and kernel-devel$(uname -r), but after that it built the guest modules correctly and seems to be running well.
I have no immediate plans for my SL virtual machine. I may use it for some coding and general goofing around with RPM. Stability is great in a long term machine for this use, and I may even switch to using it instead of my Lubuntu 12.10 VM.
And of course, a screenshot. 🙂
Having recently converted a family member to abandoning Microsoft’s products for GNU/Linux, I was faced with the problem of exporting email from Outlook ’07 to a format that would at least be text editor readable in GNU/Linux.
MS Outlook stores your email in a .pst file, which appears to simply be a proprietary archival format. You can locate the .pst file by going to the Account Settings.
I found a very handy little utility called readpst in the Debian repositories. I downloaded it, and per some instruction/manpage perusal, I ran
readpst -D -M -b -o destination_folder Outlook.pst
which unpacked the PST folder into a file hierarchy properly corresponding with the way it had been set up in Outlook. The -D option includes deleted messages, the -M makes sure the proper folder structure is maintained and attachments are placed in the right folder, the -b option specifies not to save attachments for the RTF formatting of the mail (make it plaintext), and the -o option specifies the target directory.
I use and very much enjoy using Claws-mail, a full featured mail client that weighs in at a fraction of the size of the de facto Linux mail client, Thunderbird. Claws mail saves all messages as plain text files in a transparent folder structure (no burying of mail and attachments in odd dotfiles or folders) and therefore all messages are viewable/editable with any text editor and extremely simple to access. HTML and image support is simply a plugin away as well.
To import the messages, I simply created new folders in the Claws-Mail program, and copied the exported mail into those folders in the file manager. Claws updated instantly, and all the mail was viewable along with attachments.
I couldn’t be more pleased. 🙂
This post will briefly detail my experience setting up the Linksys NP100 PCMCIA 10/100 Ethernet adapter in DSL. It is currently fully functional, and getting networking (even ethernet) on this old beast was a real benefit.
First, I was saved from the trouble I had with TinyCore by the fact that DSL run cardmgr, a utility for interfacing with PCMCIA cards. I plugged in the card and it recognized it at boot immediately- dmesg showed
eth0: Asix AX88190: io 0x300, irq 3, hw_addr 00:04:5A:A5:66:08
This card uses the AX88190 chip, which means that although normally it tries to use the pcnet_cs driver, you must manually bind it to axnet_cs.
Open /etc/pcmcia/config with a text editor, and search for the paragraph on the Linksys NP-100. Change the line that says
Next open /etc/pcmcia/config.opts, and add this to the end of the file:
card "Fast Ethernet 10/100 PC Card" version "Network Everywhere", "Fast Ethernet 10/100 PC Card", "3.0", "AX88190" manfid 0x0149, 0xc1ab bind "axnet_cs"
After this you should blacklist the pcnet_cs driver by opening /etc/modprobe.d/blacklist, and adding the line
Save and exit.
Now because DSL doesn’t autoconfigure the network (no dhcpcd or dhclient) at startup you need to set it up manually, with a static IP address:
sudo ifconfig eth0 desired_ip netmask 255.255.255.0
Changing desired_ip to your desired static IP.
This got the card functioning for me. If anyone else has any different experience or corrections to make, please leave a comment or send me an email at firstname.lastname@example.org.