SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
Play 3D Hardware Accelerated Games in Style! Also, something you can do that's really cool with Linux and simply not possible with Windows is that you can have Cedega run in a window on your desktop! Running cedega by itself will pull up a dialog. Look under "Edit->Global Settings->General" and set Desktop to something other than "no". If your using a version of cedega earlier than 5.0 or wine, look for a config file or .wine file thats normally hidden off your home directory. The best way to find this setting is to do a file text search looking for a "Desktop" = "No" or "Desktop" = . Dropline gnome users can use their Places->Search for Files option at the top of the screen to do this. You might have to change the video resolution inside your game to match what you specify here to keep them in sync. Change it to "Desktop = "800x600" or "Desktop = "1024x768" and then play a nice game running a host of other apps in your other desktop workspaces.
Also, if you do this, you'll need to change this "DXGrab" = "Y" to this "DXGrab" = "N" otherwise the game will grab your mouse (and keep it).
To change the mouse grabbing setting, the latest cedega (6.0.2) installs it's configuration file as:
In a perfect world...
In a perfect world, you just pop in your Slackware 12 Installation Disk #1 and press the button that says "Have at it!". And you can choose to live in a world where you have two partitions; one called root '/' and the other called swap 'swap'. The Slackware Linux installer will ask for this information as part of the install process. This can work fine... ...if you live in a perfect world where things never go wrong!
In the real world...
In the real world, your going to want to take into consideration things like:
How much space you have available on your hard disk(s)
Whether or not you want to share your installation with a Windows install
Whether or not you want certain benefits available from different technologies.
Security, Recovery and Performance concerns
And thats just for starters.
Now it's possible to write a book on all that but I'm not going to at this point. Suffice to say I'm going to layout my partition configuration for you and based on what your system has and on what you want to do, feel free to pick and choose as you like. I'll just indicate what I did and why!
(Currently set at 8 GB)
There used to be a time when I would section off a 512 MB partition for the root alone but those days are past. My current root partition is 8 GB in size but only because it now contains the /usr & /opt partitions plus others. This is now the preferred approach given the convenience of having just one partition to handle what is essentially the Linux operating system as deployed & updated by the providers of Slackware.
Installation of the swap partition
Code:
/dev/hda6 swap swap defaults 0 0
(Currently set at 1 GB)
Linux requires a swap partition (but can do without it if it really has to). A general rule of thumb is a 1 to 1 for every MB of Ram memory currently installed in your computer. Mind you if you have 1 GB or more your swap will generally not get used. This is a good thing. If your suffering with memory below the 1 GB borderline then expect your swap to get used more often than not.
Installation of the /tmp partition
Code:
none /tmp tmpfs defaults 0 0
(Currently sets itself at 768 MB shared memory)
Because I have 1.5 GB of ram and do not wish to have to deal with a messy /tmp directory, I choose to have a virtual /tmp partition setup that gets reset with every reboot. By default this setup will allocate 768 MB however rarely ever does more than 10-20% ever get used between reboots (I turn my machine off every night). If you know for a fact that your going to be using your /tmp partition for temporary files larger than 1/2 of your available memory, go ahead and setup a standard /tmp partition as you word a regular disk partition ( "/dev/hda3 /tmp ext2 defaults 1 1"). I would recommend however that you make it of type ext2as you should be able safely delete everything in your /tmp directory without adverse effects on whatever your doing. Remember, a /tmp is for temporary files, treat it like that and you'll have much fewer problems there!
Installation of the /shm device partition
Code:
tmpfs /dev/shm tmpfs defaults 0 0
(Currently sets itself at 768 MB shared memory)
This is a special partition used by graphics cards to share memory (shm probably means shared-memory). It rarely take up more than 1% of actual memory so don't worry about that. Just it there if your using an ATI based hardware acceleration card (whether they've asked for it or not).
Installation of the /mnt/win2k partition
Code:
/dev/hda1 /mnt/win2k ntfs-3g defaults,force 1 0
(Currently at 6.5 GB)
Chances are you already have a Windows partition on your system. Slackware will ask during the install if you would like to keep track of this and have access to it while in Linux. Typically Lilo will also allow you to boot up into Windows if you like, giving you dual boot capabilities. If you are using any sort of graphical partitioning software and wish to partition your hard disk before installing Linux, simply resize your existing Windows partition to about 5 or 10 GB. Thats more than enough for Windows. As your installing Dropline Gnome with the intention of using that for 99.99% of your basic computing needs. You just want to keep Windows there as a backup should something go wrong and you need to get to the Internet or just want to play games that won't work properly using Wine or Cedega.
Also note to use NTFS partitions, Slackware comes with basic NTFS read-only software, if you wish to read & write to the Windows partition (a must if you wish to do a Windows Partition Lobotomy) you'll need to install Fuse & NTFS-3G (described elsewhere in this thread).
Installation of your /var partition
Code:
/dev/hda5 /var reiserfs defaults 1 2
(Currently set at 512 MB)
In my setup, I've created a 512 MB partition for /var. This is helpful as the /var is where Linux will store log files and other information that changes during the actual execution of a Linux session. You can access your logs (unless secured otherwise) in the /var/log directory. All subdirectories appearing off /var follow that principle. Having a dedicated partition for /var reduces the amount read/write activity occurring in the / root directory. This reduces the chances your root partition will be damaged in the case of a power outage.
Installation of your /usr/local partition
Code:
dev/hda7 /usr/local reiserfs defaults 1 2
(Currently set at 4 GB)
You kinda want to setup a /usr/local partition even if your not a programmer. Typically, it is the /usr/local directory where you'll be placing all your special refinements and local compiles of packages & upgrades that you'll be downloading yourself. Typically this is where your Fuse & NTFS-3G packages will be compiled and installed for example. The /usr/local directory acts like an override from it's /usr counterpart. That is, if you look under your /usr/local directory you'll see things like /usr/local/bin, /usr/local/opt, /usr/local/sbin these in turn will be accessed by the environmental PATH variable before the default /usr/bin, /usr/opt, /usr/sbin found under /usr directory. This is a good thing such that if for whatever reason you are unhappy with the newly compiled package, you can safely remove it knowing that the original installed version is still available and ready to go!
Installation of your /usr/local/opt partition
Code:
/dev/hda8 /usr/local/opt reiserfs defaults 1 2
(Currently set at 6 GB)
Ditto as described for /usr/local except that this is where I install all my major packages like JBuilder, Mysql, Tomcat, Eclipse as oppose to /opt.I've given up installing major packages into /opt as I would rather leave /opt for Slackware to use as it sees fit. I'll place my optional packages inside of /usr/local/opt and have the added benefit that if I ever upgrade to say Slackware 13 at a later date, I won't have to worry about reinstalling these packages as they'll already be there and not overwritten by the new install.
Installation of your /home partition
Code:
/dev/hda9 /home reiserfs defaults 1 2
(Currently set at 13 GB)
Give your /home partition a place to call it's own. With all the remaining space on your hard disk. It's not a bad idea to make this partition the last partition as if you need to make a special partition later, you can take some off the top of this one. It's the /home partition where all your user directories will be stored.
Use fdisk -l as root to find out what you have as usb devices connected to your system. Once you get gparted installed, you can use it to graphically create partitions on those devices if you wish or just make use of what is already there.
These are part of every Slackware installation I've ever made. They are supplied by the Slackware install process and I'm not fully sure what they do but you need these specified in your /etc/fstab.
Left alone, your default Slackware/Gnome install will store everything it gets to your hard disk as soon as it gets it and do so for future reference. Things like image thumbnails however can quickly eat away at a lot of your systems resources. And while it might seem nice to have these kicking around, your system is no more immune to performance issues than Windows is if you don't address the situation properly. By declaring these directories as temporary shared memory folders your system will automatically dump temporary files with every reboot. Saving you wasted resources and possible problems arising from file corruption in the case of malfunctions hidden in HTML files. (Been there, bought the T-shirt) For every user, create a /home/{username}/.thumbnails entry (and so forth)
A special note to preferred file system type
You may have noticed a preference to using the reiserfs file system. For me, that file system type has proven to be most practical, fast and excellent in power-failure recovery. If for whatever reason your system goes down without being shut down properly, you can count on reiserfs to perform a very fast transaction based check of writes made to the partition during the next boot up. In the case of usb devices you may have to do that manually (ex. fsck.reiserfs /dev/sdb5). But it has a performance rating of 99.99% failure-recovery, not bad for free software!
Also note, that to install more than four partitions on your hard disk you'll have to setup an extended partition. When you do this, it's a good idea to let your Windows partition occupy hda1 and ideally let your root linux partition hang out at hda5. I happen to have my root at hda3, thats just how it worked out on this latest install. With an extended partition you can have any number of extra partitions but it's really not practical to have more than what's really necessary. The big reason you use partitions is so that you can take advantage of multiple file system types as well as a countermeasure to file corruption.
Hope this helps.
- Perry
Special Note to Dropline GNOME users
By default, the Dropline GNOME installer will use a directory called /var/cache/dropline-installer to place all of it's tgz files for it's installation. It's necessary to have enough space for this directory before you install Dropline GNOME as it requires 512 MB - 1 GB. As my /var directory only maintains 512 MB for its own uses, you will need to redirect this directory by doing what is called a symbolic link. To do this execute the following before attempting to run the Dropline GNOME installer!
Then proceed with your Dropline GNOME installation. Predominantly this has more to do with good organization than anything else. /usr/local/opt is what I've setup for adding 3rd party packages and Dropline is a beautiful add-on. In the future, should I make an full upgrade to say "Slackware 12.1" I won't have to worry about Dropline downloading it's packages again, as I'll have a directory in place containing the latest updates!
The scary part about all this is that Dropline already comes with Beryl ready to go, you just have to turn it on...
TODO: Turn on Beryl effects and describe the process...
But I have bit of a challenge in my case as the latest intel from ATI says that my trusty 9600 is not going to be able to provide Composite support. You however, may be able to do so.
Perry, your post is very informative and helpful, it ought to be a sticky for Slackware12, that said can you expand or provide a link expanding on your reference to the function of Extra Partitions to use Multiple File Systems. Do you mean if I a building a file basic hm server/webserver for windows and apple laptop, I should install Samba on its own partition, Apache on its own, AppleTalk on its own, vfat and then have a separate partition for all the files?? I am reading loads of how to Install and config Samba but haven't come across anything mentioning needed File Types. I want to use your guide to reinstall slackware12 but File Types is part of prep for pre-install. Be advised the network is for my use only to start off.
Step 1: Download the driver
Ok, here's the latest way to download and install your 3D hardware accelerated DRI if you own an ATI 9600/9700 + series of a video card.
Step 2: Install the driver
As a result of the last command two tgz packages will be created, both have to be installed and should be named similiar to the following:
Step 3: Config the driver
First time installers of their ATI card may also have to run the following to update their xorg.conf. But it's always a good idea to backup your current xorg.conf before you do that!
Code:
su
cd /etc/X11
cp xorg.conf xorg.conf.bak
aticonfig --initial -f
Step 4 : Have a backup plan (optional)
If in doubt, change your inittab from a 4 to a 3!
Code:
vi /etc/inittab
look for a line that says :4: in it
Code:
# These are the default runlevels in Slackware:
# 0 = halt
# 1 = single user mode
# 2 = unused (but configured the same as runlevel 3)
# 3 = multiuser mode (default Slackware runlevel)
# 4 = X11 with KDM/GDM/XDM (session managers)
# 5 = unused (but configured the same as runlevel 3)
# 6 = reboot
# Default runlevel. (Do not set to 0 or 6)
id:4:initdefault:
This will cause your reboot to go into text/console mode where you can enter "startx" to see if your system took your changes to xorg.conf. If for whatever reason your having trouble do the following:
Code:
cd /etc/X11
cp xorg.conf.bak xorg.conf
And then change your 3 back to a 4 when your ready and you'll be back where you started.
Mind you Ctrl-Alt-Bksp restarts your server, with Ctrl-Alt-Fn (1-7) will give you a chance to sign on in text as root. In which case you can enter init 3 & init 4 to restart the X server should you get a weird "pause" effect that the ATI driver will sometimes spit at you. ATI keeps trying to say they've "corrected" that problem... whatever!
Step 5: Update the kernel
Finally, should you ever make any changes to your kernel after this, it'll be a good idea to do this after you've done a make install:
Code:
(current cd in /usr/src/linux)
make install
cd /lib/modules/fglrx
make_install.sh
This ensures the fglrx driver is still part of your kernel.
If making changes to your kernel is a frequent process, make a change to your /sbin/installkernel file to speed the process up for you!
Code:
# The -p is for the few people that know you can set passwords on lilo that
# will then be required if someone wants to be sneaky and root the system
# with "linux single" or similar lameness..
/boot/build
/lib/modules/fglrx/make_install.sh
/sbin/lilo -p
Step 6: Reboot & Test
Hope this helps get you these kind of results:
Code:
perry@slackware:/mnt/maxtor/installations$ glxgears
15743 frames in 5.0 seconds = 3147.260 FPS
19605 frames in 5.0 seconds = 3920.955 FPS
18246 frames in 5.0 seconds = 3649.192 FPS
perry@slackware:/mnt/maxtor/installations$
perry@slackware:/mnt/maxtor/installations$ fglrxinfo
display: :0.0 screen: 0
OpenGL vendor string: ATI Technologies Inc.
OpenGL renderer string: ATI RADEON 9600 Series
OpenGL version string: 2.0.6747 (8.40.4)
perry@slackware:/mnt/maxtor/installations$
Step 7: Enjoy your new 3D hardware accleration
Be sure to undo the changes made for your backup plan (if you did that), change the inittab entry from 3 to 4. Mind you, I'm back to 1280x1024 resolution again (as oppose to 1400x1050). Maybe I'll make another posting on how to switch between these two drivers later!
Cheers,
- Perry
ps.
While not an *absolute* requirement, sometimes ATI requests you have a device called 'shm' supported in your /etc/fstab:
Code:
tmpfs /dev/shm tmpfs defaults 0 0
It rarely takes up much memory, it won't hurt to have it included!
(found a couple kernel options here that i'm tinkering with but only because my cedega just does not want to interface with my ati, my glxgears tells me i got hardware acceleration on an amd64, cedega says it just can't find it... wasting my time on cedega i guess... will let you all know how it turns out)
Update: it knocked everything out so bad... it was time to re-install!
AMD Athlon 64, VIA K8T800, Slackware 12, ATI Users
The HyperTransport Bus of the Athlon 64
In case you are of the above description and want to know a little more about what it is that you got underneath the hood (hyper-transport, over-clocking, etc), check out this article!
Quote:
The Basics
One of the main differences between the Athlon 64 and 64 FX processors and all the other processors in the market today is that they have an internal circuit called memory controller. In the other processors there is no such circuit, and it is the chipset of the motherboard (more specifically, a circuit called north bridge) that performs the communication task with the RAM memory. The Athlon 64 accesses the RAM memory transferring two data per clock pulse, making its maximum transfer rate with the RAM memory be 3,200 MB/s, no coincidence in the fact it is the same maximum rate of the DDR400/PC3200 memories. In other words, to make the most of this type of processor this type of memory is required.
The Athlon 64 has two external buses. One to access to the RAM memory, and the other to access the chipset. This second bus is called HyperTransport. Theoretically this architecture is better, since there is only an external bus in the other processors, which is used to communicate with the chipset (the north bridge circuit), which is responsible for the communication both with the RAM memory and with the other circuits of the computer. In theory, the Athlon 64 can communicate with the memory and with the other circuits of the computer at the same time, something impossible in the other processors, for there is only one communication way.
Another advantage of the HyperTransport is that it has a path for the transmission of data and another one for its reception. In the traditional architecture used by the other processors, a single path is used both for the transmission and for the reception of data. In theory, the Athlon 64 can transmit and receive data to the chipset at the same time.
With Slackware, at times you need a good Rescue CD - VectorLinux has yet to let me down.
They have a Live CD SOHO product with is actually based upon the Slackware engine! With it, you can boot a smaller but really really nice and easy to use Slackware distro that will allow you to inspect your main Slackware installation and if required, undo whatever it is you did that is stopping you from being able to reboot! Among other things. By itself, it's a great standalone and a really great way to try before you buy (time investment-wise that is, seeing how most of this stuff is free).
Quote:
Introduction: Speed, performance, stability -- these are attributes that set VectorLinux apart in the crowded field of Linux distributions. VectorLinux is a lighterweight, fast, Linux operating system for Intel-AMD x86 compatible systems and is based upon Slackware, one of the original Linux distributions. Slackware is the true 'Unix' of Linux distributions and its popularity stems from the fact that it is a robust, versatile and almost unbreakable system. VectorLinux has improved Slackware to produce a bloat free, easy to install, configure and maintain operating system that is second to none. We include automatic hardware configuration, unique administration tools and easy software package management via the Gslapt/slapt-get system. VectorLinux is considered to be the fastest, non-source Linux distribution on the planet!
Give VectorLinux a try and discover the difference........ Welcome home.
Make it a point to burn a Vector Linux Live CD (SOHO is a good choice) and have it next to your computer should you need it. You'll be glad you did!
What does a marine, a porn star and a Linux developer have in common?
Ans: "They're always playing around with their tools!"
When to do a fresh reinstall!
Honestly, if only I could bill myself out on how much time I spend tinkering around with my Linux install only to *tinker* too much to where I have to turn around and fix the very same problems I help create, I'd be rich!
There comes a time when all you want to do is a complete reinstall, when this happens, be sure to backup your /etc/fstab, /etc/lilo.conf and your /etc/X11/xorg.conf (if you can) to give you a head start on your next install.
Using /usr/local & /usr/local/opt in their own partitions
If your smart, you would have placed all your added packages and all your work area into these folders which should have had their own partitions. If this is the case, then you can safely drop your root partition (as I described above) and you'll find that you've preserved 99.99% of everything you were working on before the re-install. If you didn't do this, either backup whats there or get ready to reinstall whats there.
Using /home in it's own partition
The same holds true for the /home folder such that you *want* to get that it's own partition. Otherwise you'll be either backing that up or throwing it all away. It's the /home directory where all your users have their workspace, personal files, that kind of stuff. By having that as it's own partition, you won't have to worry about loosing anything when re-install. Be careful when you recreate your users however. Done properly you can recreate your users with the same identifiers and values so that they can access their previous home directories as if nothing had happened. Sometimes things get messy and it's best to create a new home directory for each user and provide them with their older home directory as a sub directory to the new one. While I'm not a huge fan of KDE, it has it's perks, kuser is the best user manager I've used so far for Linux. If you do this, synchronize the user and his or her old files with the chown & chgrp commands.
Reinstalling is not a bad idea! Sometimes it's necessary, other times your just wanting to re-assemble your rifle. After all, if it's something you enjoy using, your going to want it in good shape. Especially, if you've been doing a lot of kernel compilations and the like. Seems odd, but based on the lessons you learned from your first installation, your 2nd will all that much smoother (provided you are in a good frame of mind). It's always good to do something like this after a good nights sleep. This minimizes the number of silly mistakes!
The rewards are great! Right now I'm looking at a brand new install and I am impressed, things that were causing me problems (like HAL not working right, the screensaver not activating reliably, the kernel not compiling properly, and a bunch of other tiny *gotchas*) all seem to be ironed out with a fresh install. So after trying to figure out why a little change in the kernel config file, might have caused a lock up, that might have caused my /var (formerly in ext2 format, bad idea) to get corrupted last night, I'm now typing to you in a much healthier system, things are looking good.
Keep track of the files you install! Along with the rescue CD I mentioned earlier, another saving grace is that I copied all the files, including the iso's into a specially designated directory where I could get access to them again rather than having to track everything down.
Code:
root@slackware:/home/perry/Desktop# ls /mnt/maxtor/installations -la
total 39436
drwxr-xr-x 4 perry users 4096 2007-08-21 21:15 ./
drwxr-xr-x 11 perry users 8192 1969-12-31 16:00 ../
drwxr-xr-x 20 perry users 4096 2007-08-29 09:25 slackware12/
drwxr-xr-x 8 perry users 4096 2007-08-18 17:59 windows2k/
root@slackware:/home/perry/Desktop# ls /mnt/maxtor/installations/slackware12 -la
total 1024
drwxr-xr-x 20 perry users 4096 2007-08-29 09:25 ./
drwxr-xr-x 4 perry users 4096 2007-08-21 21:15 ../
drwxr-xr-x 4 perry users 4096 2007-08-21 22:26 ati/
drwxr-xr-x 3 perry users 4096 2007-08-29 09:20 beryl/
drwxr-xr-x 2 perry users 4096 2007-08-29 09:25 cmos/
drwxr-xr-x 3 perry users 4096 2007-08-06 15:56 dropline/
drwxr-xr-x 4 perry users 4096 2007-08-06 16:51 etc/
drwxr-xr-x 3 perry users 4096 2007-08-07 11:03 flashplayer/
drwxr-xr-x 3 perry users 4096 2007-08-29 09:18 gdesklets/
drwxr-xr-x 2 perry users 4096 2007-08-06 16:43 gkrellm2/
drwxr-xr-x 2 perry users 4096 2007-08-29 09:19 gparted/
drwxr-xr-x 3 perry users 4096 2007-08-06 13:26 how-to/
drwxr-xr-x 3 perry users 4096 2007-08-04 17:33 ignore/
drwxr-xr-x 3 perry users 4096 2007-08-29 23:07 kernel/
drwxr-xr-x 5 perry users 4096 2007-08-21 22:29 lq.org/
drwxr-xr-x 2 perry users 4096 2007-08-05 10:36 ntfs-3g/
drwxr-xr-x 2 perry users 4096 2007-08-29 09:18 python/
drwxr-xr-x 4 perry users 4096 2007-08-05 22:11 slackware/
drwxr-xr-x 3 perry users 4096 2007-08-06 15:36 swaret/
drwxr-xr-x 4 perry users 4096 2007-08-29 22:34 wireless/
-rwxr-xr-x 1 perry users 962604 2007-08-02 20:16 simplylinux.txt*
With that directory in place, putting a fresh install together was pretty much like reassembling my rifle after I cleaned everything. Keep these little tips in mind, you'll be rewarded with a highly responsive, fun to use computer that's not laden with viruses, spyware, malfunctions, poor optimizations or other problems!
Today's hardware just keeps getting faster, cheaper, better! Wouldn't it be a shame to have all the speed and firepower of your graphics card and hard driver performing no better than the computer you had 10 or 20 years ago because of poor maintenance? Can you remember, yea old Commodore Vic 20? (it has a "real keyboard!")
Duke Nuke'em to the rescue!
Sometimes it can get pretty hectic making a change to a kernel config and then dealing with the consequences. You need a fallback in case your latest change did things like knock out the mouse. When this happens you need to apply a *divide-&-conquer* approach to isolating or backtracking. Assuming you have been able to boot into your system at all, have the following files already placed in your /boot directory:
The tgz files of the best Slackware kernel packages that worked for you!
The following build script
The following nuke-m script
the kernel packages
Code:
cd /boot
ls *.tgz
kernel-generic-smp-2.6.21.5_smp-i686-2.tgz
kernel-modules-smp-2.6.21.5_smp-i686-2.tgz
The nuke-m script will allow you, with the least amount of effort to quickly reinstall the last Slackware distributed kernel and it's accompanying modules into your system thus putting your system back into useable condition. This also isolates a probable source of error, this way you know that nothing you did in your kernel config is preventing your system from being able to reboot. A very handy piece of information in light of how easy it is to make a change to your BIOS (thus causing erratic behaviour all over the system) and think it's your kernel when really it's something else entirely! This is why we call the technique *divide-&-conquer*.
Please note: This is a *Tactical* Nuclear Strike Only! And not a thermal nuclear war! There is a difference! In the case of this tactical nuclear strike, your only changing back your kernel vmlinuz file (and associated links) and it's modules. You are not undoing that 45 minute compile of the kernel that's still in place. Those files are waiting for you to undo that last change to the kernel config so that you can (relatively) quickly *unfix* what you *remember* doing and then do another make. That's pretty much what happened to me this morning, so doing a nuke-m will only get rid of the problem in the last compile vmlinuz file and not wipe out your *huge* compile, thus allowing another famous program called *make* to do it's job. Provided your change wasn't a change in processor type, your recompile should only take a couple of minutes to fix. And then your on your way with your kernel tinkering a little bit more relaxed now that you have what might be called your *exit* strategy!
The sweet side of all this is that I now have a very smoothly running system, that just feels so very light weight and well under control. One of the many benefits of tinkering with something for a while. You just get better at it... some glad I don't do drugs!
If for whatever reason you've knocked out your lilo, mbr or lilo.conf enough so that your system can't boot off the hard drive, simply pop in your Slackware #1 installation CD and execute only it's Configure option under setup.
Setup will ask you where your partitions are (that's a logical thing to do) and later it'll give you the option of overwriting your previously setup /etc/fstab. It's generally NOT a good idea, not to overwrite a perfectly working /etc/fstab as you may have placed many refinements there that simply will not come to mind while 1/2 in a panic. Your task here is to re-create your /etc/lilo.conf not your /etc/fstab. Setup just needs to know where to find the /etc/lilo.conf file. So when it asks if you'd like to re-write the /etc/fstab file, it's usually better to say *no*.
Setup's Configure will ask a few other questions, as long as one of them isn't do you want to reinstall all your packages or format anything, your fine.
This saves you the panic of having to reinstall everything from scratch just because of one little file (aka lilo.conf) being knocked out.
I guess so many years of using computers and you sort of develop this notion that a 64-bit processor operating at 3000+ MHz, 1.5 Megabytes of ram and memory operating at 400 Hz, plus almost 500 Gigabytes of hard disk space and just maybe... MAYBE you can get a cool & quiet performance setup out of it.
How I pity anybody depending on Vista or even XP on one of these machines as all those resources just kinda get eaten up. Somehow, some way all that hardware goes to squat with their so-called operating system software.
Here in the free world, we get to install our brand of Linux an d tonight it's paying off for me as I have this smooth running machine under Slackware 12 / Dropline and I honestly cannot believe at how well behaved it's being. With all those bells and whistles running, everything in harmony, who would ask for more!
It's almost 12 pm and hence time for bed, but I'd rather stay up and mall my Linux-ed computer some more, see what else I can tweak!
Just had to share that with someone, hope your computer is treating you good!
Taking the time to setup for yourself a good backup plan is a smart idea. Should your main kernel in your boot directory ever get corrupted, or (more likely) a kernel compile didn't go so good, you'll need a way to get back into your machine easily!
As you can see (below), I am making use of the same kernel that is used by Slackware for installation purposes. As it is intended for emergency use only, it's purpose is to allow me to get back into my system and undo any changes that caused the /boot kernel to fail. Typically followed by a /boot/build (to regenerate the initrd.gz) and a lilo. Whenever you make changes to your kernel setup you must call lilo to let the system know where to find the latest kernel on bootup.
Note to self: /boot/build & lilo
The last three commands to execute after installing your latest kernel upgrade and before you reboot should always be:
Code:
cd /boot
./build
lilo
Here's my /etc/lilo.conf:
Code:
# LILO configuration file
# generated by 'liloconfig'
#
# Start LILO global section
boot = /dev/hda
compact # faster, but won't work on all systems.
prompt
lba32
timeout = 50
# Normal VGA console
vga = normal
# VESA framebuffer console @ 1024x768x64k
# vga=791
# VESA framebuffer console @ 1024x768x256
# vga=773
# ramdisk = 0 # paranoia setting
# End LILO global section
# Linux bootable partition config begins
image = /boot/vmlinuz
initrd = /boot/initrd.gz
root = /dev/hda3
label = slackware
read-only # Partitions should be mounted read-only for checking
vga=773
# Linux bootable partition config ends
# Linux bootable partition config begins
image = /boot/huge/vmlinuz
root = /dev/hda3
label = slackware.huge
read-only # Partitions should be mounted read-only for checking
vga=773
# Linux bootable partition config ends
# Linux bootable partition config begins
image = /test/vmlinuz
root = /dev/hda3
label = slackware.test
read-only # Partitions should be mounted read-only for checking
# Linux bootable partition config ends
# Windows bootable partition config begins
other = /dev/hda1
label = windows
table = /dev/hda
# Windows bootable partition config ends
Just the bare basics but complete with backup plan in case of emergency.
NOTE: image = /boot/huge/vmlinuz
I copied the contents of /boot and placed them into /boot/huge. Then inside /boot/huge I redirected vmlinuz, config and System.map to point to the settings for the huge (non-smp) kernel! The reason for doing so is to minimize any negative impact the latest smp compile may have on the main kernel in /boot. This gives me two backup options if there is something wrong with the most recently compiled kernel. The first one being /boot/huge, the other being /test! The /test directory is an exact copy of the original /boot directory, placed in it's own area so as to minimize problems that may occur should either the /boot or /boot/huge directories get mangled.
Compiling your kernel with less to worry about
This arrangement gives me the ability to compile both the kernel and it's modules with minimal concern should something not agree with the hardware. With two backup options, plus the benefit that the latest compile being able to go into the system with minimal fuse, I can customize the kernel with ease!
Just a tip: there is no need to create additional directories just to have a second (or 3rd, etc...) kernel entry in /etc/lilo.conf
That kernel in lilo.conf does not need to be called "vmlinuz". The "vmlinuz" name is a convenience symlink just like all the other symlinks in /boot - it makes a Slackware install and kernel upgrade easier to script. You can however just use the real names in /etc/lilo.conf and keep all your kernel stuff nicely in one and the same /boot directry.
"Activate your defense matrix!" - C&C'95
In case your wondering, Slackware 12 + Dropline comes all setup to play DVDs on your system provided you have a DVD drive. You will have to install an additional library to read encrypted DVDs however:
Code:
su
wget http://www.kenzalewski.com/Slackware-11.0/libdvdcss-1.2.9-i486-1kjz.tgz
installpkg libdvdcss-1.2.9-i486-1kjz.tgz
'/mnt/dvdrom not found'
For reasons unknown, this may come up the moment you pop in your dvd. To correct this:
Code:
su
mkdir /mnt/dvdrom
Last call for alcohol! The following relates to versions of Slackware earlier than 12 but if your having trouble getting your dvd drive accessed properly, try the following:
Quote:
How do I add a DVD player?
Well to start you will need a /dev/dvd after you boot watch the bootup process to make sure what device it is or (dmesg at console). In your case as you said hdd. Make sure /dev/dvd points to /dev/hdd it should be a symlink if not create it "ln -s /dev/hdd /dev/dvd" man ls You need to make sure DMA is turned on for the device. "hdparm /dev/dvd" will display if it is or not. If it is not set then issue a "hdparm -d 1 /dev/dvd" hdparm --? The player should be ready to use after this. Just pick your fav player mplayer xine etc.. and watch your movie.
Last update: 2005-09-16 21:53
Author: Jim Simmons
Revision: 1.0
You should also be able to access the file directory on the DVD and copy them as files!
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.
"Want to play a game of chess?" - HAL
Configuring Linux Partitions
What does a marine, a porn star and a Linux developer have in common?
