NFS lock problem

Rob Landley rob at
Mon Mar 20 16:14:18 UTC 2006

On Monday 20 March 2006 4:33 am, Aubrey wrote:
> Hi all,
> I'm working on the blackfin uclinux platform. When I mounted the nfs
> as the root file system, the syscall "fcntl" can lock file
> successfully. But when I start up kernel and mount(busybox) the nfs on
> a local directory, the syscall "fcntl" can't lock the file. The error
> log as follows:
> ===================
> <5>Jan  6 06:06:29 kernel: lockd: cannot monitor
> <5>Jan  6 06:06:29 kernel: lockd: failed to monitor
> <7>Jan  6 06:06:30 kernel: nsm_mon_unmon: rpc failed, status=-13
> ===================
> It seems the lockd wasn't started. But the mount parameter was:
> ===================
> on /mnt type nfs
> (rw,v3,rsize=32768,wsize=32768,hard,udp,lock,addr=
> ===================
> I didn't use the "nolock" option, So I thought the lockd should be
> started by busybox-mount command.
> Is it a bug of busybox or I'm wrong?

I don't use nfs.  I take it this is a userspace daemon that busybox doesn't 
contain, but which busybox is expected to find and spawn?  (I thought your 
init scripts were supposed to start the various security-hole daemons 
associated with nfs.  I know Red Hat's do because stopping the suckers was 
always one of my first steps cleaning up a Red Hat system back when I ran 

Never bet against the cheap plastic solution.

More information about the busybox mailing list