| Previous | Table of Contents | Next |
After your machine reboots, the LILO prompt should appear. Make sure that you can boot to your old operating system if you left it on the hard drive. If that system was DOS, press <Shift> and then type the short word you used to identify the DOS partition when you installed LILO. If you enter an invalid word, press <Tab> to get a list of valid operating system types. If you’re having problems at that point, place your DOS boot disk in the boot drive and reboot.
You should be able to boot from your boot disk. When your system is up and running under DOS, try the Linux boot disk you created during installation—not the ones you created to originally install the entire system. If that boot disk doesn’t work, you may have to reinstall Linux. Potential problems to check initially are the kernels and your hardware. Before starting over, make sure that you have the appropriate hardware. If you made notes during the installation process, check which kernel you installed against what hardware you have. Make sure your hardware is supported by Linux.
Below are some answers to common problems listed on Red Hat’s Web site. These troubleshooting tips are used under the provisions of GNU’s GPL.
Q: Can I use a hard drive that has more than 1023 cylinders?
A: The infamous 1023 cylinder question. Yes, but not to boot Linux. You can install Linux on partitions above the 1023 cylinder, but to boot Linux, the root directory and specifically the /boot directory must be installed on the first hard drive below 1024.
Q: How do I add arguments for LILO at the prompt?
A: Some hardware requires that extra parameters be fed to the kernel before the kernel will recognize the hardware. You can accommodate this by editing the /etc/lilo.conf file to provide the necessary parameters, or you can provide them manually during boot up. See the LILO HOWTO for more examples of LILO parameters.
Q: Why does LILO hang on LI?
A: This is a symptom of the 1023 cylinder problem addressed previously. If you have installed the boot system above 1023, LILO will not be able to boot the system. You can try to boot from a floppy using the rescue disk you made during installation, or you can repartition your hard drive and reinstall Linux.
Q: The installation will not find the SCSI card.
A: To remedy this, you need to add a boot-time argument such as the following:
LILO: linux qlogicfas=0x230,11,5
This option can be made permanent so you don’t have to re-enter it. See the LILO configuration option append in the lilo.conf man page.
Q: How do I uninstall LILO?
A: If you want to uninstall LILO and reinstall the original boot record, try using this command
lilo -u /dev/hda
which represents the boot record of the first IDE drive. Parameters may vary for your machine, for example, if your first hard drive is a SCSI drive, you would use /dev/sda.
Q: Can I use LILO and Win95 on one installation?
A: Yes, install Windows 95 first and then install Linux. During the installation, tell Linux to place LILO in the MBR. You can also use a commercial program such as System Commander.
Q: How do I mount a CD-ROM?
A: Installing Red Hat 5.1 should place the proper entries in your /etc/fstab file, as shown below:
# # /etc/fstab # # You should be using fstool (control-panel) to edit this! # # <device> <mountpoint> <filesystemtype> <options> <dump> <fsckorder> /dev/sda1 / ext2 defaults 1 1 /dev/sda5 /home ext2 defaults 1 2 /dev/cdrom /mnt/cdrom iso9660 noauto,ro 0 0 /dev/fd0 /mnt/floppy ext2 noauto 0 0 /dev/sda6 /var ext2 defaults 1 2 /dev/sda2 none ignore 0 0 0 none /proc proc defaults /dev/sda7 none swap sw
Note the use of noauto for the cdrom entry. Without this setting, Linux will try to automount the CD-ROM when it boots, which isn’t really a problem unless there’s no CD in the drive.
If there is not an entry in your fstab file, you can either edit /etc/fstab or use the X Window Control Panel tool to add the appropriate mount information. Also, make sure the mount point /mnt/cdrom does indeed exist. If the entry is correct, you can cd to the mount point and issue the following commands:
cd /mnt mount cdrom
Q: I have Red Hat 5.0 and have upgraded to the ld.so RPM package listed in the errata, but my libc5 applications still create a seg fault. What is wrong?
A: The problem with crashing libc5 applications can be caused by several things.
Before or after the upgrade, another version of libc might have been installed that was not obsoleted by the upgrade process, or the libc5 libraries might have been placed in a location that causes conflict.
To find out if this is the case, run this command:
rpm -qa | grep libc
It should produce the following output:
glibc-devel-2.0.5c-12 libc-5.3.12-24 glibc-debug-2.0.5c-12 rpm-2.4.10-1glibc rpm-devel-2.4.10-1glibc glibc-profile-2.0.5c-12 glibc-2.0.5c-12
If you see items like libc-debug-5.3.12-18 or libc-5.4.44-2, you will need to remove these packages (with the command rpm -e libc-debug ) and run ldconfig –v.
Your /etc/ld.so.conf file has been changed from an optimal setting. For optimal loading, set your /etc/ld.so.conf file in the following order:
/usr/i486-linuxaout/lib /usr/i486-linux-libc5/lib /usr/openwin/lib /usr/X11R6/lib
Q: Some of my older applications get the incorrect time.
A: Some libc5 apps want /usr/lib/zoneinfo. You can either recompile them for libc6, or you can provide a symlink with the following command so things will work.
ln -s ../share/zoneinfo /usr/lib/zoneinfo
Q: I have all the latest updates installed, but my programs still get the incorrect time.
A: If you have installed all the latest updates and your programs still get the incorrect time, try checking the settings in /etc/sysconfig/clock. They probably look something like this:
| Previous | Table of Contents | Next |