check, if /dev is mounted. the advantage is, that you can crypt the whole disk only using a kernel and an initrd. problem: if /dev is mounted with the needed /dev/mapper/*-information for fsck,... the information gets lost when "overmountig" the new ramfs. Reproducible: Sometimes Steps to Reproduce: for my crypt-disk i
check, if /dev is mounted. the advantage is, that you can crypt the whole disk only using a kernel and an initrd. problem: if /dev is mounted with the needed /dev/mapper/*-information for fsck,... the information gets lost when "overmountig" the new ramfs. Reproducible: Sometimes Steps to Reproduce: for my crypt-disk i´m using devfs on the pre-init-process and later udev. after initialising the /dev/mapper/* with cryptsetup i copy /dev/* to the new /dev-dir. theoretically this is only necessary the first bootup, cause on shutdown /dev is saved and on bootup restored. but, what will happen, if you update or change your filesystem..... workaround: in /sbin/rc comment try mount -n -t ramfs none /dev out. this is not nice....
set RC_DEVICES in /etc/conf.d/rc to 'static'
???? RC_DEVICES is never used in my /sbin/rc if [ "${udev}" = "yes" ] then ebegin "Mounting ramfs at /dev" # new code if [ -z "$(cat /proc/mounts | awk '{ print $2 $3 }' | grep devramfs)" ] then try mount -n -t ramfs none /dev eend $? fi # end new code # i do not like this if-statement ####### try mount -n -t ramfs none /dev ####### eend $? ebegin "Configuring system to use udev" if [ "${RC_DEVICE_TARBALL}" = "yes" ] then einfo " Populating /dev with device nodes..." try tar -jxpf /lib/udev-state/devices.tar.bz2 -C /dev fi populate_udev if [ -e /proc/sys/kernel/hotplug -a -x /sbin/hotplug ] then einfo " Using /sbin/hotplug for udev management..."
you're using baselayout 1.9.x upgrade to 1.11.7-r2 or better
cool service :D - this was very fast.... thx. ill try it tomorrow ;) .... now its too late
