when I tried Marc`s patch for edgy, I stumbled across a major security issue. When usplash exits early (as it does for example on a fsck on my machine) the entered password is echoed plain text on the console. So I think it`s not such a good idea to attach to a console when usplash is running.
I then made altered the first few lines of cryptdisk.functions in the following way:
pgrep usplash
if [ $? -gt 0 ]; then
USPLASH_RUNNING="false"
else
USPLASH_RUNNING="true"
fi
The attached patch is a modified version of Marc`s patch, that adds these changes.
Now it works perfectly for me using LUKS volumes (a quick test showed that other dm-crypt devices are still ignored when usplash is active).
Hi,
when I tried Marc`s patch for edgy, I stumbled across a major security issue. When usplash exits early (as it does for example on a fsck on my machine) the entered password is echoed plain text on the console. So I think it`s not such a good idea to attach to a console when usplash is running.
I then made altered the first few lines of cryptdisk.functions in the following way:
pgrep usplash
if [ $? -gt 0 ]; then RUNNING= "false" RUNNING= "true"
USPLASH_
else
USPLASH_
fi
stdin=`readlink /proc/self/fd/0` /dev/null} " != "$stdin" ] && [ "$ON_VT" != "yes" ] && [ "$USPLASH_RUNNING != "yes ]; then
if [ "${stdin#
exec env ON_VT=yes /usr/bin/openvt -f -c `fgconsole` -w $0 "$@"
fi
The attached patch is a modified version of Marc`s patch, that adds these changes.
Now it works perfectly for me using LUKS volumes (a quick test showed that other dm-crypt devices are still ignored when usplash is active).