Description of problem:
misconfiguration about rpc.statd options between /etc/sysconfig/nfs and /etc/rc.d/init.d/nfslock

Version-Release number of selected component (if applicable):
nfs-utils-1.0.9-35z.el5_2

How reproducible:
always

Steps to Reproduce:
1. configure /etc/sysconfig/nfs with
STATDARG="-n nfsserv1"

2. start nfslock service

3. ps -ef | grep rpc.statd | grep -v grep
  
Actual results:
rpcuser   2063     1  0 08:57 ?        00:00:00 rpc.statd

Expected results:
rpcuser   2063     1  0 08:57 ?        00:00:00 rpc.statd -n nfsserv1

Additional info:

Currently if you want to pass arguments to rpc.statd, you have inside
/etc/sysconfig/nfs
# Optional arguments passed to rpc.statd. See rpc.statd(8)
STATDARG=""

and infact I set up it like this:
STATDARG="-n nfsserv1"

But inside /etc/rc.d/init.d/nfslock start section you have instead:

        if [ -n "${STATD_HOSTNAME}" ]; then
        STATDARG="-n ${STATD_HOSTNAME}"
        else
        STATDARG=""
        fi

so that if you set STATDARG in etc/sysconfig/nfs, it is cleared by nfslock init script.
So at the moment the solution is to not use STATDARG inside /etc/sysconfig/nfs and instead add the line 

STATD_HOSTNAME=nfsserv1

But if you want to add more options to rpc.statd you have problems.
Probably the correct way is to delete completely the section about STATD_HOSTNAME in /etc/rc.d/init.d/nfslock and incorporate it directly in STATDARG globally in /etc/sysconfig/nfs (as outlined in the comment of the file itself, see above).

Note that the problem is still present in latest F9 x86_64 too:
nfs-utils-1.1.2-6.fc9.x86_64