These are the most frequently asked questions about ZipSlack: Q1: How do I add X to this? Q2: I get "unable to open virtual console" when I boot! Q3: Why does the boot process stop with this "Kernel panic" error? Q4: I can't unzip the file -- it says I don't have enough memory! Q5: All I have is DOS! Do you know of an unzipper that works with zipslack.zip for DOS or Windows 3.1? Q6: How can zipslack.zip be split into floppy sized chunks? Q7: Can my ZipSlack installation be moved onto a real Linux partition? -------------------------------- Q1: How do I add X to this? A: Assuming you've got the space to install it, download the X packages (the files ending in .tgz) from this directory: ftp://ftp.cdrom.com/pub/linux/slackware/slakware/x1/ You might also grab the diskx1 file, which describes each of the packages. You can save disk space if you don't install all of the X servers (you only need the one for your video card), and you might also leave out extra fonts, old X shared libraries (oldlibs*.tgz), and possibly development tools if you're not planning to compile X software yourself. Once you've collected the packages you plan to install into a directory, run this to install them: installpkg *.tgz Before running X, you'll need to set it up with 'xf86config'. -------------------------------- Q2: I get "unable to open virtual console" when I boot! A: You're probably not giving LINUX.BAT the correct partition name. If you really have no idea which one to use, you can try each of these. If it's on an IDE partition, it will almost definately be one of them: /dev/hda1 /dev/hda2 /dev/hda3 /dev/hda4 /dev/hda5 /dev/hda6 /dev/hdb1 /dev/hdb2 /dev/hdb3 /dev/hdb4 /dev/hdb5 /dev/hdb6 /dev/hdc1 /dev/hdc2 /dev/hdc3 /dev/hdc4 /dev/hdc5 /dev/hdc6 /dev/hdd1 /dev/hdd2 /dev/hdd3 /dev/hdd4 /dev/hdd5 /dev/hdd6 If you know which hard drive (not C:, D:, etc, but which number drive, 1, 2, 3, or 4) the partition is on, then you can narrow the list down quite a bit. The first IDE drive's partitions all start with /dev/hda, the second hard drive's partitions begin with /dev/hdb, and so on. This can also be caused by unzipping zipslack.zip in the wrong place on a partition. It must be unzipped in the top directory on the partition (such as in the C:\ directory) or the kernel won't be able to find the installation when it boots. The unzipping process will create a new directory for the files automatically (C:\LINUX). -------------------------------- Q3: Why does the boot process stop with this "Kernel panic" error? VFS: Cannot open root device 08:04 Kernel panic: VFS : Unable to mount root fs on 08:04 A: Well, you see, device 08:04 is the fourth partition on a SCSI drive (/dev/sda4). In most machines these days, there are no SCSI devices, only IDE ones. (NOTE: the parallel-port version of the Zip drive *is* treated as a SCSI device by Linux) What you need to do is edit the LINUX.BAT file and change the uncommented loadlin line (without the 'rem') to boot the partition where you installed Linux. This is probably something like /dev/hda1 if you used your C: drive. If you have no idea what the partition is, look right before it stops and you'll see something like: hda: hda1 hda2 hda3 In this case, the partition must be one of: /dev/hda1, /dev/hda2, /dev/hda3. -------------------------------- Q4: I can't unzip the file -- it says I don't have enough memory! A: This happens if you use a 16-bit unzipper. You must use a 32-bit version such as WinZip or PKZIP for Windows95 or NT. (of course, unzip for Linux also works fine :) -------------------------------- Q5: All I have is DOS! Do you know of an unzipper that works with zipslack.zip for DOS or Windows 3.1? A: Yes, (for Win 3.x) I hear there's something called Zip Navigator. If you can't find that, here's how to unzip it under Linux. First, grab a bootdisk that works with your system. For an average IDE system: ftp://ftp.cdrom.com/pub/linux/slackware/bootdsks.144/bare.i For a SCSI system: ftp://ftp.cdrom.com/pub/linux/slackware/bootdsks.144/scsi.s For an IBM PS/2 microchannel bus machine: ftp://ftp.cdrom.com/pub/linux/slackware/bootdsks.144/ibmmca.s If you're unzipping the file to a parallel port Zip drive: ftp://ftp.cdrom.com/pub/linux/slackware/bootdsks.144/iomega.s Then, you'll need the rescue disk. This is a small Linux system on a floppy disk. One of the utilities it includes is unzip. Here's where you can get the rescue disk image: ftp://ftp.cdrom.com/pub/linux/slackware/rootdsks/rescue.gz Boot the first disk (the bootdisk), and hit enter at the 'boot:' prompt. When prompted, insert the rootdisk (rescue.gz) and hit enter to load it. Log in as root. Now, you'll need to mount the DOS partition where the zipslack.zip file resides. If the final destination is a Zip disk, you might still just want to go ahead and unzip the file on your DOS partition, and then move the directory tree onto the Zip disk under DOS with XCOPY or a similar tool. If you're not sure what the name of your DOS partition is under Linux, use this command: fdisk -l | more The partition name will be something like /dev/hda1. To mount the partition, use a command like this: mount /dev/hda1 /mnt -t msdos Switch to the /mnt directory, and unzip the file: cd /mnt unzip zipslack.zip That should unzip the file. Now you're ready to hit crtl-alt-delete and reboot your machine. From there, follow the directions in README.1st to boot the Linux OS. -------------------------------- Q6: How can zipslack.zip be split into floppy sized chunks? A: Splitting up the file is another way to handle the problem of 16-bit unzippers failing to unzip ZIPSLACK.ZIP. Here's a report from one of our users about how to do this: I had trouble unzipping - I don't have a win95 machine, and pkunzip for DOS croaked horribly. Info-zip unzip did a much better job, but died after about 3/4 of the unzip process (out of memory). So, I used a "zip splitter" to split the zipslack.zip into floppy sized bites (from Simtel, filename zc300.zip), then extracted each of the 26 zips to the Zip disk. Booted like a champ, runs nice but slow on a 486/66 with 12Mb ram. -------------------------------- Q7: Can my ZipSlack installation be moved onto a real Linux partition? A: Yes, it can. Here are the steps you'll need to follow to migrate your installation onto a Linux ext2 partition: 1. Define a Linux partition using fdisk or cfdisk. If you find it easier, you can use DOS or Windows tools to create the partition and then use Linux fdisk to change the partition type to 83 (Linux native). 2. Format this partition with 'mke2fs'. For example, if your new Linux partition is /dev/hdb1 you'd use the following command: mke2fs /dev/hdb1 Formatting destroys the existing filesystem on the partition, so make sure to format the correct partition! 3. Mount the new partition on /mnt. In the case of the example above, this command will do it: mount /dev/hdb1 /mnt 4. Make a few directories on the new partition: mkdir /mnt/cdrom mkdir /mnt/mnt mkdir /mnt/proc 5. Now it's time to actually move the data. First you'll need to set your 'umask' to 000 to correctly preserve all file permissions, and then you'll copy the top-level directories (other than cdrom, mnt, and proc) and the kernel file (vmlinuz) onto the new Linux partition: umask 000 cp -a /bin /mnt cp -a /boot /mnt cp -a /dev /mnt cp -a /etc /mnt cp -a /home /mnt cp -a /lib /mnt cp -a /root /mnt cp -a /sbin /mnt cp -a /tmp /mnt cp -a /usr /mnt cp -a /var /mnt cp -a vmlinuz /mnt If you've made any new top-level directories that you want to save, copy them over to the new partition in the same way. 6. Edit the /mnt/etc/fstab. Change the device listed for the '/' partition to the new Linux partition's device. 7. That's it! Your system should be ready to boot on the new partition. To do that, you can use loadlin (if it's installed on your DOS or Windows partition), or a bootdisk. Once you've booted the new partition you can proceed to set up LILO if you like. Note that when you boot a native Linux partition you should boot it in read-only mode (unlike UMSDOS). This allows it to do automatic filesystem checking periodically, or if the machine is ever shut down improperly. To boot a partition in read-only mode, add 'ro' instead of 'rw' to the bootdisk or loadlin command line. Since this operation must be done as root and involves dangerous operations like using fdisk, you need to be careful to avoid losing data. But, if you can migrate your installation successfully, you've earned your intermediate Linux sysadmin merit badge. :^) --------------------------------