FreeBSD Manual says host can't modify files on the NFS unless -maproot specified in /etc/exports "If the -maproot=root flag is not specified, then even if a user has root access on the remote system, he will not be able to modify files on the exported file system." First of all, this only means the client can't have root permissions on the NFS server. It's deceptive because if -maproot-root is NOT specified, the root user on the client does have ability to create and save files on the NFS. I learned that doing so (being root on the client & writing files to the NFS) I see that the file's owner is 4294967294 and group wheel. (This # is what results when -2 is coerced into a non-signed integer on a 32-bit system). In other words, the root user on a client is given -2 (priority) on the NFS server and not a "0" that the root ususally enjoys. Of coure, the user on the client has to SU just to mount the NFS. Then exit out of the SU and provided the user is named "kat", writing files to the NFS will show a file owned by "kat" instead of it being owned by 4294967294.
As I understand it, NFS deamon gives access to machines via the /etc/exports file. So, why isn't there a way to have the client SU and mount a NFS and then return back to a user account without having a simple user be told how to mount an NFS on their laptop?
As I understand it, NFS deamon gives access to machines via the /etc/exports file. So, why isn't there a way to have the client SU and mount a NFS and then return back to a user account without having a simple user be told how to mount an NFS on their laptop?