Red Hat Bugzilla – Bug 177781
mkinitrd init script leaves /loopfs mounted read-write
Last modified: 2007-11-30 17:11:20 EST
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8) Gecko/20051201 Fedora/1.5-1.1.fc4.nr Firefox/1.5
Description of problem:
I am booting Fedora from a root filesystem image file that is stored in an ext3 filesystem on /dev/hda1. mkinitrd generates an init script that handles all the loopback magic just fine.
The only problem is that the init script leaves the filesystem containing the root image mounted read-write on /loopfs. When I try to mount the same filesystem after the system finishes booting, fsck has to do a journal recovery because the filesystem is dirty.
/proc/mounts shows the filesystem still mounted as /loopfs. However, when I try umount /dev/hda1, I get "not mounted".
Version-Release number of selected component (if applicable):
Steps to Reproduce:
Run mkinitrd with / configured as /dev/loop0, and the loopFs stored on an ext3 (or any writable) filesystem. /etc/fstab looks like this:
/dev/loop0 / ext3 defaults 1 1
# LOOP0: /dev/hda1 ext3 /boot/rootfs
/dev/hda1 /disk ext3 defaults 1 1
Expected Results: The init script should mount the loopfs read-only, since the only thing it needs to do is read the root filesystem image file.