I'm trying to mount an NFS drive, but I am getting an RPC Authentication Error, with the reason given as that the credentials are too weak.
On the server side (192.168.100.109), it looks like the NFS server and mount daemon are starting and being available to the target client (192.168.100.101):
# cat /etc/exports /datastore/winmedia -ro 192.168.100.101 # portmap # nfsd -tun 4 # mountd # showmount -e 192.168.100.109 Exports list on 192.168.100.109 /datastore/winmedia 192.168.100.101
However, when trying to mount the disk on the client-side, I receive an RPC Authentication Error
# ls /mnt/winmedia # mount -v 192.168.100.109:/datastore/winmedia /mnt/winmedia mount: no type was given - I'll assume nfs because of the colon mount: RPC: Authentication error; why=Client credential too weak
Looking on the server logs, it only says that the connection was refused, but doesn't give a reason.
Feb 11 23:43:25 fattyopenbsd mountd: Can't open /var/db/mountdtab: No such file or directory Feb 11 23:48:41 fattyopenbsd mountd: Refused mount RPC from host 192.168.100.101 port 62826
On the client side,
rpcinfo shows the same data as on the server side, except without the program names.
Server side rpcinfo:
# rpcinfo -p 192.168.100.109 program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100003 2 tcp 2049 nfs 100003 3 udp 2049 nfs 100003 2 tcp 2049 nfs 100003 3 udp 2049 nfs 100005 1 udp 630 mountd 100005 3 udp 630 mountd 100005 1 tcp 612 mountd 100005 3 tcp 612 mountd
And on the client side, same thing but no process name
# rpcinfo -p 192.168.100.109 program vers proto port 100000 2 tcp 111 100000 2 udp 111 100003 2 tcp 2049 100003 3 udp 2049 100003 2 tcp 2049 100003 3 udp 2049 100005 1 udp 630 100005 3 udp 630 100005 1 tcp 612 100005 3 tcp 612
So, my questions are 1) Why is the mount being refused? (Is there a way to get more detailed information about the failed request?) 2) How can I make the NFS drive be mounted?
The server is running OpenBSD 5.8
The attempting client is Puppy Linux, and is a VirtualMachine on a Windows machine. (Not sure if the VM IP is an issue. If I run
ifconfig on the client, it shows a
10.0.*.* address on eth0, but the Windows host has the 192...101 address over wireless. I figured since the server log shows the request coming from the ...101 address, the VM issue is probably not an issue, but if I am wrong any suggestions would be great).
UPDATE I now believe this issue is related to the VMs network settings. The request to the server is coming from a Non-privileged port number ( >1024) on the VM's host machine.
The VM is using NAT, so even if the VM uses a privileged port when making the request, this is not reflected in the final request.
This should be fixable by using a bridge connection to the host instead of NAT, which would cause the requesting IP and port to be the same as the VM, but sadly I have only wifi at my place and couldnt get the bridged connection working with wifi adapter.
Solution: use Samba instead of NFS, so that a username and password are used for authentication instead of uid, ip address, and client side port number.