Comments on Setting Up An NFS Server And Client On Debian Etch

Kerberos should be used, encryption is important and Userauth. The other Thing is, NFSv4 got now the Port 2049, this makes it ease to work with firewalls, than with the old one.
/home 192.168.0.101(rw,sync,no_root_squash)
This makes it ease to steal this share. So kerberos would be better.
Diffrences between rsize and wsize are absolut important, imo, the new one makes 32768 the old one 8192.

Greetings Seraphyn

It should be noted that in Debian 4.0 (Etch) the portmap utility may not install properly (I think there is something in the configure portion of the package that is broken).  This may in turn cause long delays for mounts to work as well as break any file locking operations over NFS.  The workaround to this is on both client and server system run:

 dpkg-reconfigure portmap

then make sure to say "no" at the 'bind portmap to localhost' question

Also, you may need to add entries on both servers in /etc/hosts.allow:

client system: 

    portmap: 192.168.0.100

    statd: 192.168.0.100

server system:

    portmap: 192.168.0.101

    statd: 192.168.0.101

Note that later versions of nfs-kernel-server will gripe at each filesystem export about a missing option in the exports file no_subtree_check | subtree_check.  I advise going with 'no_subtree_check' as an option added to each export.  This will of course vary with the operations you expect to be performed on the NFS share.  

After several rather unsuccessful attempts get nfs working through diverse books and websites finally a comprehensive and easy to follow tutorial about nfs.

I have installed nfs4 in my Debian Server 5.0.3 with kernel 2.6.18.8, my problem is port 2049 didnt LISTEN so the client nfs didnt connect to the server, here the information :

ns2:/srv# rpcinfo -p
program vers proto port
100000 2 tcp 111 portmapper
100024 1 udp 33284 status
100024 1 tcp 39745 status
100000 2 udp 111 portmapper
100003 2 udp 2049 nfs
100003 3 udp 2049 nfs
100021 1 udp 33325 nlockmgr
100021 3 udp 33325 nlockmgr
100021 4 udp 33325 nlockmgr
100005 1 udp 33326 mountd
100005 1 tcp 51781 mountd
100005 2 udp 33326 mountd
100005 2 tcp 51781 mountd
100005 3 udp 33326 mountd
100005 3 tcp 51781 mountd

ns2:/srv# lsof -i:2049
it show nothing!
ns2:/srv# nmap localhost

Starting Nmap 4.62 ( http://nmap.org/ ) at 2009-11-05 07:43 UTC
Interesting ports on localhost (127.0.0.1):
Not shown: 1702 closed ports
PORT STATE SERVICE
21/tcp open ftp
22/tcp open ssh
25/tcp open smtp
53/tcp open domain
80/tcp open http
110/tcp open pop3
111/tcp open rpcbind
143/tcp open imap
443/tcp open https
953/tcp open rndc
993/tcp open imaps
995/tcp open pop3s
3306/tcp open mysql

ns2:/srv# telnet localhost 2049
Trying 127.0.0.1...
telnet: Unable to connect to remote host: Connection refused

ns2:/srv# /etc/init.d/nfs-kernel-server restart
Stopping NFS kernel daemon: mountd nfsd.
Unexporting directories for NFS kernel daemon....
Exporting directories for NFS kernel daemon....
Starting NFS kernel daemon: nfsd mountd.

ns2:~# tail -f /var/log/messages
Nov 5 06:52:21 ns3 kernel: nfsd: last server has exited
Nov 5 06:52:21 ns3 kernel: nfsd: unexporting all filesystems

Hi,

I would like to know how do we enable NFS access logs by client ip addresses. 

The reason being, there is an infected Desktop in our LAN network, and it dumps the worms on the NFS share on Iomega appliance which based on Debian 5.0

Is there a way of finding out, which desktop is the culprit?