Hi,...
Ok this might sound really stupid, but I've seen several ways of doing
this and want to get this cleaned up.
Nevertheless....
kernel-parameters.txt has the following options where you can specify
the root-fs and related flags:
root= [KNL] Root filesystem
nfsroot= [NFS] nfs root filesystem for disk-less boxes.
Plus further options like: ro, rootdelay=, rootflags=, rootfstype=,
rootwait, etc.
My question:
These are intended (especially root/nfsroot) to really name the device,
on which the filesystem directly lays, right?
E.g. if the ext4-root-fs is on /dev/sda1,... => root=/dev/sda1
I've seen several initramfs scripts (which I'd like to fix),... where
people abuse this or misuse this in setups where a root-fs is used on
multi-stacked block layers (e.g. something like disk->lvm->dm-crypt->fs)
Cheers,
Chris.
--
