In late 2006 I stumbled upon a bug in which anaconda would throw an exception if a %post script changed /etc/mtab to be a link to /proc/mounts. I had assumed this was due to anaconda's attempts to update /etc/mtab as it unmounted filesystems before it exited. I even went as far as to strace anaconda to verify it indeed was mucking around with /etc/mtab after the %post script ran. However, I went to reproduce (and then fix) this problem today, but couldn't reproduce it. Does anyone know for sure if this was fixed? Or should I dig deeper to see if there's another element necessary to trigger the failure? (the original kickstart that triggered this failure had a lot of other stuff in its %post that perhaps could play a role in triggering the failure). Jeff
