So what is all this linux crap I hear about?

Terminal - there are some good prompter sheets around - it's a massively powerful tool but it does take some getting used to. Try these:

http://fosswire.com/post/2007/08/unixlinux-command-cheat-sheet/

http://fosswire.com/post/2008/04/ubuntu-cheat-sheet/

 

http://ubuntuforums.org/forumdisplay.php?f=93

The forum there might have something on World of Warcraft. As usual try a word search and if nothing found then register and ask, they're usually pretty helpful.

- - - Updated - - -

Yes there is a steep learning curve with Linux but it is worth the bother. Good thing is that there is plenty of info on the net and you will find a lot of fast help if you need it.

 

WoW is a very common "I need help" area. There should be an ubuntu forum which walks you through it start to finish... I have never run it. I will figure it out and get back to you.

Any other questions you have... feel free to ask. I will address them as quickly as I can. Remember, almost any command in a terminal has a -h or --help associated with it, as well as a man page.

All of your system information is available via the subsystem. So to combine these two ideas... try the following.

lspci -nn | grep VGA

This will return your gpu information. On this laptop for instance:

This tells me, obviously, that I am running an ATI Radeon HD 3200... but there are some other interesting bits of information here. 1002=ATI... all manufacturers have a unique ID. 9612 is the specific chipset pciid... all devices have their own unique number as well. 01:05.0 is a hex representation of the physical bus the device is attached to.
cat /proc/bus/pci/01/05.0

Neat huh? What you are seeing there is firmware code. Clearly not English.

So we used 3 commands here.
lspci = list pci devices. See also ls for files/directories (works like dos "dir") lsmod for modules (drivers) currently loaded into the kernel, lsusb for listing usb devices recognized, and more rarely lspcmcia for those slot devices that fit into laptops. You will see I put -nn after the lspci, because I wanted those pci numbers -nn basically means give me any and all numbers that can be returned. This is called a switch or an argument... in that it produces some kind of change for the information being delivered. Just running lspci will produce basic information on everything attached to the pci subsystem. Try lspci -vvn this produces verbose output with numbers on everything.

grep = grep is one of the most powerful tools in the arsenal. It returns matching patterns from anything. You can grep for a word or series of words recursively for any output or file containing that pattern. You should note the pipe | ... probably the most flexible tool ever. I piped the output of lspci to grep... telling it I only want it to return the result containing the pattern VGA. Try lspci -nnk | grep VGA -A3

We told lspci to show us numbers and kernel module information, then grepped VGA again, this time telling it also to return the 3 lines that come after the initial return. We can pipe outputs over and over again to new and different commands. So... for instance...
lspci -nn | grep VGA | while IFS=':. ' read -r tok1 tok2 tok3 rest; do printf '%2s %2s %s\n' "$((16#${tok1}))":"$((16#${tok2}))":"$((16#${tok3}))" | sed -e 's/ //g'; done

Now... that is a little more complicated. What it does is take that bus assignment and spit it out as a base 10 number that xorg would like to see, if you were going to hardcode the appropriate bus in there (say for instance you have more than one GPU showing and wanted to designate an output to one).
This isn't a very impressive example... just turning 01:05.0 into 1:5:0... however it is also converting from hex to a base 10 system... you just don't notice it here because of the numbers being used. You see the first part of that is what we used to get the line earlier describing my video hardware. Instead of doing that, I am going to specify an output to show conversion.

echo "01:14.d" | while IFS=':. ' read -r tok1 tok2 tok3 rest; do printf '%2s %2s %s\n' "$((16#${tok1}))":"$((16#${tok2}))":"$((16#${tok3}))" | sed -e 's/ //g'; done

The last command we used was cat. You use cat to read files... what it actually does is redirect the contents of a file to standard out (your monitor).
cat /etc/issue will tell you information about your system.
cat /proc/cpuinfo will tell you more than you want to know about your processor.
grep -i model /proc/cpuinfo will return information with "model" in it... the -i means ignore the case.

So what I want, is the model name... so I adjust my command.
grep 'model name' /proc/cpuinfo | sort -u

So what is different here? I have added single quotes around my search term, because it has a space in it, which grep would be expecting a location after. It is a literal string I am looking for here... so I use single instead of double quotes. You will also see that I piped it to the sort command. -u means return only unique results. This is "how" to use the command line. Mostly what you will be using it for is apt (from aptitude), which is a very popular package management system for dpkg (the debian package manager). apt-get install gdmap would install the utility gdmap. Want to know what gdmap is?
apt-cache show gdmap

Want to know what version is available, and if you have it installed?
apt-cache policy gdmap

apt-cache search gdmap would return any packages which have anything to do with gdmap... gdmap is little so it would only return it... however if you did a search for xorg you would return LOTS of results for packages that have something to do with your display.

apt-get update, updates YOUR SYSTEM to know the AVAILABLE PACKAGES... it does not actually change any of your software.
apt-get upgrade will actually upgrade all of your software to the latest available release on a given distribution.
apt-get dist-upgrade will upgrade your kernel to the latest, and do a regular upgrade considering the new kernel.
apt-get do-release-upgrade will bump you up a release version. I stick to long term service distributions... so I don't use this much.

Don't fear the command line. The CLI demonstrates the power of your fully operational battle station!!!

So to address your actual want...
http://www.mmomeltingpot.com/2012/0...ing-agent-exe-failed-and-other-update-errors/
http://www.wowwiki.com/World_of_Warcraft_functionality_on_Wine

These two pages should provide the information/patches/tweaks needed to get it going. It is a shame Blizzard abandoned Linux users. There was a time when they were doing beta on linux... the porting shouldn't be difficult at all as it runs on mac... they just... don't.

 

Thanks for all the info. That's a lot to study on. I have been working on my Wow problem since posting last, and I think I've got it fixed. None of the forums I saw said exactly what to do in a way that worked exactly like they said it would, but I got enough bits and pieces of info to improvise something which appears to have done it. Can't be sure yet b/c the World is down for maintanence, but I actually have the program up, rather than just the laucher, so I think I'm good. Thanks.

 

The coolest thing about the shell (Born Again SHell in your case)... is that once you know how to get the information you want, and issue commands based on that information... you know how to program. You just dump all the information you want into a file that begins with a shebang and the shell type... so... lets take some of the information we gathered, and some new stuff and make a script that tells us about our computer.
#! is a shebang. It is special. Anything else beginning with a # means "ignore this line from here on"

Now you can put that in a file somewhere... and in the CLI, navigate to it.
ls ~ will list files/folders in your home directory... which is typically located at /home/username. If you created it in your Documents directory, then ~/Documents will be its location

You should become familiar with the laziest most awesome tool ever... tab complete. Linux wants to complete your desires for you... so if you type: cd ~/Doc and hit tab... it will complete Documents for you. If you had a folder Documents and a folder Doctors it would not complete anything, but if you hit tab again, it will show you both options so you can add another letter so it can complete. If you type ifc(tab) it will complete to ifconfig... because that is the only command you have avail beginning with ifc etc. It does this with variables and... really anything you can think of. It is very smart.

So... we need to make this script executable. Lets pretend you named this file talkit.
sudo chmod +x ~/Documents/talkit

This will make it executable. Because we are doing systemish things, this script will have to be run with sudo privelege as well.

sudo ~/Documents/talkit

The first time it is run it will install espeak... then, it will tell you (audibly) all sorts of information about your computer.

You can run anything inside of $( ... ) as standalone commands from above to see what they are returning if you are curious.

So... that is how you do shell programming. Pretty cool.

 

If I install this stuff, what will happen to my windows? will it still be there if I cant figure Linux out? will I still be able to find all my files? fear is the reason we don't try it because MOST of us really have no idea how all this stuff is working behind the scenes.

 

If you do a side by side install, it will produce a grub menu when you boot and you can choose which you would like to boot into.


Again... if you just want to try it out without committing to an install... you can run it live from the install disk (on distros with a live disk of course).

You just pop it in the cd/dvd/usb slot and boot to the cd/dvd/usb device... then you can choose to install it, or run it. You can fool around quite a bit from the live environment. You cannot install TOO much stuff, because you only have so much buffered room in ram to pretend to be a hard drive... and when you reboot, anything you installed or changed will be gone. (This assumes you have not set up persistence on the writable medium such as a USB key) Also keep in mind that you are running everything off of a CD effectively, so performance will be much slower than an installed version... though probably faster than windows

 

don't have the CD and don't plan on looking for it but thanks for your advice

 

I am referring to the Linux install CD, you would have to download it, and burn it to optical or USB thumbdrive, and either boot linux from there live, or install it. If you go with a *buntu derivative, it will detect and suggest installing along side windows. It will repartition/size the drive and install itself next to windows, and give you an option to boot to whichever you like whenever it starts.

I would recommend going here... and downloading the LTS (Long Term Support) version. It will come as an .iso file. This is a complete cd/dvd image. Your cd burning software should be able to burn the iso image (not the file) to a blank cd or dvd... but I recommend using USB. Much faster and less error prone.
http://www.kubuntu.org/getkubuntu

Get a usb thumbdrive, then go here and download the universal USB installer
http://www.pendrivelinux.com/universal-usb-installer-easy-as-1-2-3/

and update it... follow the instructions. This will install 12.04 onto the USB drive.

When your computer boots, most have a quick message telling you to hit an F key, typically f12, to choose the boot device. Choose the thumbdrive. If you do not have such an option you will have a BIOS option (could be f1 f2 esc del etc). Change the boot order there.

Then feel free to install or just try it out live.

 

thank you my good man. I will try and get my nerve up! lol

 

np...

Note: arm8+ architecture can come with a "Windows only" UEFI that cannot be disabled.

This is gay.

RedHat has purchased a key.
http://linux.slashdot.org/story/12/...l-pay-microsoft-to-get-past-uefi-restrictions

 

Here comes my Cliff Notes after Ctrl's post-grad lecture

1. If you decide to install on the HD then you will find most Linux distros, particularly the *buntus, are kind enough to ask you if you want to install on the whole HD or alongside Windows. I find that a blessing.
2. As Ctrl has pointed out, the boot menu gives you a choice, hop in and select the one you want to boot at any given time and off you go.
3. It works really well. I find that I can get into the Windows files from Ubuntu (and Linux Mint which is Ubuntu-based) but Windows will not let me get into the Linux files (just as well because I'd probably cause merry hell in there).
4. Backup.
5. Good luck with it, you'll soon use Windows for games and possibly business (I have to use Office for business and it's a fine product, LibreOffice is great but not quite what I want especially with PPT which I use a lot). Dual/triple boots rule!

 

I have heard about this. This sounds to me like a monopolistic practice that makes their Internet Explorer fiasco look totally kosher by comparison. Any idea how much of the new hardware has this bull(*)(*)(*)(*)?

 

It occurs to me, it is bad practice to run strange scripts you don't quite understand from strangers with root privileges... so instead, install espeak on your own (sudo apt-get install espeak) and then run that script as a normal user by invoking it at its location... ~/Documents/talkit in the case of the example I gave.

 

All Windows 8 licensed certified hardware... (is the goal anyway... def all win cert ARM 8+)
http://www.zdnet.com/blog/open-source/linus-torvalds-on-windows-8-uefi-and-fedora/11187

Not for the squeamish.
http://www.techrepublic.com/blog/opensource/hardware-neutrality-uefi-strikes-again-and-again/4150

 

I wouldn't stake much on this claim. Is it free from majority of the ridiculous amount of security holds that Mac or Windows produces? Absolutely. It remains to be known that most of the viruses, even on Macs & Windows, come from ill-advised users who insert the malicious code onto their computers by accident (IE: Opening up an attachment in an e-mail).

In your defense, I'd say the reason behind why Linux is least populated with viruses is because the users are a lot smarter. This is no way, shape, or form a bash to the overall intelligence to those who don't use it (I use Windows & Mac and would at least believe I can hold myself on solid footing around other geeks), it's merely a point that most folks probably don't even remember when Windows was actually MS-DOS. We now consider the 'command prompt' as an addition to the OS when, in fact, it was the premise of the Windows OS in the beginning. Let alone that we've become so graphically oriented (Even with video games), most folks would rather not type in commands to do the work. They'd rather 'double click' on something and let the OS take over from there.

However, another geekish point, Microsoft is now pushing very hard for PowerShell. They are trying to compete with UNIX/Linux with it, we'll see how well they do with that, considering how fatally flawed the Windows OS is.

 

Opening an attachment on an email will not infect Linux. This is a difference you are dismissing and attributing to smarter users. It is not. It is a fundamental difference in security structure. I like everything else you have said.

 

Correct sir. I mis-worded my statements in such a way that would lead folks to believe that was the case. I was merely trying to state that most viruses come from ill-advised users, on Macs or Windows computers, allowing the malicious code to be inserted into their system, one example being opening up an attachment in an e-mail. I wasn't trying to imply Linux could have this happen.

However, depending upon the type of attachment & system you're using (IE: .rpm attachment on an RPM-based system), a virus could be inserted into a Linux system through an e-mail attachment. Although, the Linux user would have to initiate the installation of the file before the virus can be installed into the OS. Whereas a Windows computer, once you double click on the file, you've initiated a program needed to run the program. Thus, releasing the virus into your computer. Again, coming down to the individual knowing what/what not to install/open on the computer. Or, at least having some sort of scanner that can be run on e-mails to let the individual know that the attachment might have malicious code within it.

Some more information on this for anyone passing by this thread: http://www.linux.com/learn/tutorials/284124-myth-busting-is-linux-immune-to-viruses

 

Correct. You would have to use root permissions to install a package file... and there is simply no reason to EVER do this unless you are dealing with developers who are packaging code for you...

You CAN infect your linux system... but you would have to be completely... COMPLETELY uninformed to do so. All the software the "normal" user needs is either in the repo, or available from hardware manufacturers (for most recent versions etc). An abnormal user, yes, you are correct, will be smarter than the average bear and not fooled by such nonsense. Thanks for the comments and discussion!

 

Absolutely. Anytime sir, I stumbled on this section of the forum by mistake. I'm at home with folks who love technology. Anytime I'm able to engage in friendly conversation with folks, the benefit will always be mutual.

 

Hi Ctrl! Nice thread!

I was going to check out Kubuntu on a Win 7 laptop. Would it work ok in Virtual Box just to see the features? Or would it be better just booting it from a mem stick?

 

So, if I understand you correctly, the only way Linux can get a virus is if you input your "install a program" password? And since you SHOULD know whether or not you are installing a program at the moment, a virus could only sneak in while you are installing something else. Is that right?

 

Almost... all correct except that sneaking in bit... if you are installing from repo, they are all MD5 crosschecked... so it isn't getting in that way... you would have to download a package from someplace shady, and then install it manually using dpkg (on a debian distro that is) with root privilege.

 

I'm curious to know if anyone has ever tried the CentOS (http://www.centos.org/). I ask because I was looking at trying out a version of Linux that was near Red Hat Enterprise Linux. From what I've dug up, it seems this version of Linux is about the closest to it as I can get, outside of trying/buying the RH version.

Thoughts?

 

I personaly don't believe there is a need for a single user going for RH Enterprise linux. You really don't get anything special out of it if you are a single user. RHE is great for a business since it is tailored for mass remote deployment and management but that is of no use for the common joe. And the productivity apps that comes with it are available on other distros. I also hate RPM based package, but that is just me...