Journaling file systems reduce the time needed to recover a file system that was not unmounted properly. While this can be extremely important in reducing downtime for servers, it has also become popular for desktop environments. This chapter contains a variety of journaling file systems.
Ext3 is a journaling file system that is an extension to the ext2 file system. It is backward compatible with ext2 and the conversion from ext2 to ext3 is trivial.
You don't need to install anything to use ext3, all the required packages are available with a bare LFS system.
When building the kernel, ensure that you have compiled in ext3 support. If you want your root partition to be ext3, then compile the ext3 support in the kernel, else you may compile it as a module. Recompile the kernel if needed.
Edit your /etc/fstab. For each partition that you want to convert into ext3, edit the entry so that it looks similar to the following line.
/dev/hdXX /mnt_point ext3 defaults 1 0
In the above line, replace /dev/hdXX by the partition (e.g., /dev/hda2), /mnt_point by the mount point (e.g., /home). The 0 in the last field ensures that the partition will not be checked for consistency during the boot process by the checkfs script. You may replace the ext3 fs type in the above by auto if you want to ensure that the partition is mounted if you accidentally skip enabling the ext3 support in the kernel.
For each partition that you have converted to ext3 in /etc/fstab, enable the journal for the partition by running the following command.
tune2fs -j /dev/hdXX
Remount the concerned partitions, or simply reboot if you have recompiled the kernel to enable ext3 support.
More information is available at http://www.zip.com.au/~akpm/linux/ext3/ext3-usage.html. This informaion is still relevant to the 2.6 kernels.