The first step to recover from a disaster is to make a custom boot disk. It is not sufficient to just copy the standard boot disk image from the Linux cd rom because it will not necessarily have all the Kernal support configured. First log into the root shell session by using the cd / command and inserting a floppy disk. Then examine the /etc/lilo.
conf to find the Kernal identity and use the mkbootdisk command and device /dev/fd0 with the string, example 2. 2. 5-15, found in the /etc/lilo. conf. The uname a command can be used to display the Kernal version.
After the successful completion of that step eject the disk and write protect it. Another way to recover from a disaster is to make a rescue disk. The rescue disk is taken from the image in the linux cdrom. It is not like the custom bootdisk where an examination of the /etc/lilo.
conf is done. What makes the boot disk created for the system valuable is that it contains the special drivers, such as SCSI drivers, that may be required to boot the system. Another alternative to recover from a disaster is to create a Toms Root Boot Diskette. The tomsrtbt distribution may be done on the internet: A complete Linux system on a single diskette.
To create the diskette set the browser to http://www. toms. net/rb/ and download the latest version. To prepare the emergency boot disk, place the zipped tar file into a directory such as /usr/local/src and follow the install procedures.
The tar file is a versatile archiving utility that is not limited to strictly tape backup. TAR files can be stored locally as files, and sent over nertworks as a method of back up. The package squeezes an amazing number of Linux utilities onto a single 1. 7 MB floppy by using high compression. ConclusionNo matter what method used to back up the Linux system it is very important to take caution when dealing with Kernal config files. Always use the man command before using a command so the right syntax and option are used so that an important file isnt deleted.www.google.comwww.linux.com